http://www.informationweek.comInformationWeeken-usCopyright 2012, UBM LLC.2012-12-03T08:00:00ZFour business scenarios illustrate how small and midsize companies can reduce costs, improve disaster recovery and more via storage virtualization.http://www.informationweek.com/storage/virtualization/storage-virtualization-gets-real/240142882?cid=RSSfeed_IWK_Authors <!-- KINDLE EXCLUDE --> <!-- InformationWeek Digital Issue--> <div id="inlineGreenPromoTop"> <div class="greenBand"></div> <div class="inlineGreenPromoContent"> <a href="http://www.informationweek.com/gogreen/120312smb/?k=axxe&cid=article_axxe_os"><img src="http://twimgs.com/informationweek/supplement/046/smallcov.jpg" alt="InformationWeek Green - Dec. 3, 2012" title="InformationWeek Green - Dec. 3, 2012" align="left" class="greenIssueImage" /></a><br /> <a href="http://www.informationweek.com/gogreen/120312smb?k=axxe&cid=article_axxe_os"><img src="http://twimgs.com/infoweek/graphics_library/misc/Green_leaf_88x88.jpg" alt="InformationWeek Green" title="InformationWeek Green" align="right" class="greenLeaf" /></a> <div class="greenPromoText"> <strong><a href="http://www.informationweek.com/gogreen/120312smb?k=axxe&cid=article_axxe_os">Download the entire December 2012 <em>InformationWeek SMB</em> special issue on storage virtualization</a></strong>, distributed in an all-digital format as part of our Green Initiative<br /><br /> (Registration required.)<br /> </div> </div> <div class="greenBand"></div> </div> <!-- / InformationWeek Digital Issue--> <br /><!-- leave as a br to not interfere w/ the insights boxes --> <!-- /KINDLE EXCLUDE --> <img src="http://twimgs.com/informationweek/supplement/046/046SUP_SMBinsideart_flat_110.jpg" width="110" height="110" alt="Storage Virtualization Gets Real" title="Storage Virtualization Gets Real" width="110" height="110" class="artInlineTopImage" /> <P> Storage virtualization can deliver benefits such as better utilization of existing storage, easier provisioning, improved performance of storage systems and applications, and lower-cost disaster recovery. But many different technologies fall under the "storage virtualization" umbrella. We'll use a fictional company to illustrate how small and midsize businesses might craft a storage virtualization strategy that meets different business needs and balances cost and performance trade-offs.</p> <P> In particular, we'll look at using cloud gateways for faster file services and host-based replication for disaster recovery. We'll show how hybrid storage that combines flash storage and traditional spinning disks can deliver faster performance while controlling costs, but that doing so means sorting through several possible approaches. And last, we'll show how server-side caching done right can improve virtual desktop performance.</p> <P> In our scenario, a new management group has taken over Acme Inc., a manufacturer of novelties and toys. Over the past several years, Acme has made limited investments in its IT infrastructure as a result of the economic downturn. The company's new CIO believes in the concept behind a software-defined data center, in which software performs functions such as networking and storage virtualization that have in the past been performed by dedicated hardware. He has asked his infrastructure group to virtualize as much new infrastructure as possible -- including the storage.</p> <P> One of the first storage applications to be virtualized at Acme was file services. Before the upgrade, Acme had traditional disk-based network-attached storage systems in its Los Angeles design and distribution center, and in its three sales offices across the country. Acme has millions of CAD and graphics files in its archive of product designs and marketing materials, and the company's designers are often kept waiting as the NAS systems struggle to deliver these large files.</p> <P> <!-- KINDLE EXCLUDE --> <div style="float:right;padding-left:10px;"> <div style="width:210px; border:1px solid #000000;"> <div style="margin:0; padding:5px; background-color:#CC0000; text-align:center; font-size:1em; color:#ffffff; font-weight:bold;"><a href="http://reports.informationweek.com/abstract/24/8697/Storage-Server/research-state-of-storage-2012.html?cid=pub_analyt__iwk_20121203" target="_blank" style="color:#ffffff;">Research: State of Storage 2012</a></div> <img src="http://twimgs.com/informationweek/supplement/046/046CSreportcover.jpg" width="175" height="111" style="margin:15px;"> <div style="font-size:.9em; margin:0px 1px 0px 10px;">Our full report on <a href="http://reports.informationweek.com/abstract/24/8697/Storage-Server/research-state-of-storage-2012.html?cid=pub_analyt__iwk_20121203" target="_blank">the current state of storage</a> is free with registration. This report includes <strong>44</strong> pages of action-oriented analysis, packed with <strong>37</strong> charts. What you'll find: <ul class="normalUL"><li>Survey results from more than 300 IT professionals</li> <li>Strategies to handle growing data and shrinking budgets</li> </ul> <center><strong><a href="http://reports.informationweek.com/abstract/24/8697/Storage-Server/research-state-of-storage-2012.html?cid=pub_analyt__iwk_20121203" target="_blank">Get This</a> And <a href="http://reports.informationweek.com/">All Our Reports</a></strong></center><br /></div> </div> </div> <!-- /KINDLE EXCLUDE --> <P> Acme also struggled with transporting files from location to location. Most of the time, employees sent files via email attachments to co-workers in other offices and to the company's Asian manufacturing partners. This clogged up the Exchange server and backup repositories with multiple copies of files. In addition, employees began bypassing corporate IT, and its security safeguards, by using consumer services such as Dropbox.</p> <P> <strong>Cloud Gateways Speed File Servers</strong></p> <P> To address these problems, Acme chose a cloud storage service that uses cloud gateways like those from Nasuni and Panzura. These gateways are deployed on premises at Acme's offices, and connect to cloud storage services from providers such as Nirvanix and Amazon's S3. The gateways in each location use local solid-state drives and spinning disks to cache actively accessed data while presenting a single integrated file namespace to the users, regardless of their location. The LA design center will have a higher-end appliance with SSDs to provide the performance the designers need, while the sales offices and manufacturing partners can use less-expensive virtual appliances running under VMware's vSphere to keep costs reasonable.</p> <P> <!-- KINDLE EXCLUDE --> <center><strong>To read the rest of the article,<br /><a href="http://www.informationweek.com/gogreen/120312smb?k=axxe&cid=article_axxe_os">Download the December 2012 <em>InformationWeek SMB</em> special issue on storage virtualization</a></strong></center><br clear="all" /></p> <!-- /KINDLE EXCLUDE -->2012-10-18T00:34:00Zhttp://reports.informationweek.com/abstract/24/8960/Storage-Server/fundamentals-virtualization-lets-you-store-more.html?cid=RSSfeed_IWK_Authors2012-08-27T08:00:00ZEven small IT shops can now afford thin provisioning, performance acceleration, replication, and other features to boost utilization and improve disaster recovery.http://www.informationweek.com/news/240006046?cid=RSSfeed_IWK_Authors <!-- KINDLE EXCLUDE --> <!-- InformationWeek Digital Issue--> <div id="inlineGreenPromoTop"> <div class="greenBand"></div> <div class="inlineGreenPromoContent"> <a href="http://www.informationweek.com/gogreen/082712smb/?k=axxe&cid=article_axxe_os"><img src="http://twimgs.com/informationweek/supplement/038/smallcov.jpg" alt="InformationWeek SMB - September 2012" title="InformationWeek SMB - September 2012" align="left" class="greenIssueImage" /></a><br /> <a href="http://www.informationweek.com/gogreen/082712smb/?k=axxe&cid=article_axxe_os"><img src="http://twimgs.com/infoweek/graphics_library/misc/Green_leaf_88x88.jpg" alt="InformationWeek Green" title="InformationWeek Green" align="right" class="greenLeaf" /></a> <div class="greenPromoText"> <strong><a href="http://www.informationweek.com/gogreen/082712smb/?k=axxe&cid=article_axxe_os">Download the entire September 2012 issue of <em>InformationWeek SMB</em></a></strong>, distributed in an all-digital format as part of our Green Initiative<br /> (Registration required.)<br /> </div> </div> <div class="greenBand"></div> </div> <!-- / InformationWeek Digital Issue--> <br /><!-- leave as a br to not interfere w/ the insights boxes --> <!-- /KINDLE EXCLUDE --> <P> Storage virtualization features such as thin provisioning, acceleration, and replication have joined basic RAID as standard features of storage systems designed for relatively large data centers. Smaller companies with modest budgets and IT teams have had to muddle along with direct attached storage and entry-level disk arrays without a lot of virtualization features.</p> <P> Not anymore. All the important storage virtualization features are available as software that can run on a standard x86 server, putting storage virtualization well within reach of small and midsize businesses. In addition, VMware and Microsoft have been putting more storage virtualization and storage management features, such as live migration and thin provisioning, into their hypervisors.</p> <P> The upshot is that companies of almost any size can use more of their existing disk space (thus putting off additional purchases), improve application response times through faster storage performance, and speed the recovery of business operations via replication if and when disaster strikes. </p> <P> <strong>On Target</strong></p> <P> A common approach to storage virtualization involves using a software target. A software target is a piece of software that turns a general-purpose x86 server and its local disks into a storage appliance. The most basic versions, like Microsoft's free iSCSI target for Windows servers, run as a service on a server and make some or all of their storage available as iSCSI volumes. Because each iSCSI LUN is a virtual disk file in the server's file system, volumes are thinly provisioned by default.</p> <P> Simple targets are useful in development or test environments, but they lack the reliability needed for production use because the server they run on has multiple single points of failure. If the processor, RAID controller, or other key component fails, the storage system is dead and data is probably lost.</p> <P> Many storage target software products, including those available from StarWind Software and Open-E, address the single-point-of-failure problem by synchronously mirroring data from the primary active storage server to a second server with its own storage and an automated failover process. This means the servers can be used in environments that require high availability.</p> <P> Another high-availability option comes from NexentaStor. While most storage target software relies on hardware RAID controllers built into the server (or added to a PCIe slot), NexentaStor is based on ZFS, a combination file system and volume manager that does RAID but also solid-state disk-based read and write caching, data compression, and data deduplication in software. High-availability options for NexentaStor include an active-active dual-node configuration with shared JBODs.</p> <P> <!-- KINDLE EXCLUDE --> <center><strong>To read the rest of the article,<br /><a href="http://www.informationweek.com/gogreen/082712smb/?k=axxe&cid=article_axxe_os">Download the September 2012 issue of <em>InformationWeek SMB</em></a></strong></center><br clear="all" /></p> <!-- /KINDLE EXCLUDE --> <P> <!-- KINDLE EXCLUDE --> <center> <div id="inlineReportPromo"> <div class="inlineReportPromo_headline"><a href="http://reports.informationweek.com/abstract/25/8960/Virtualization/fundamentals-virtualization-lets-you-store-more.html?cid=pub_analyt__iwk_20120827" target="_blank" style="color:#ffffff;">Fundamentals: Virtualization Lets You Store More</a></div> <div class="inlineReportPromo_inner"> <center><strong>Virtual Storage, Real Options</strong></center><br /> <img src="http://twimgs.com/informationweek/supplement/038/038SUP_SMB_CSreport.jpg" width="175" height="115" style="float:right;"> Our full report, <a href="http://reports.informationweek.com/abstract/25/8960/Virtualization/fundamentals-virtualization-lets-you-store-more.html?cid=pub_analyt__iwk_20120827" target="_blank">"Virtualization Lets You Store More,"</a> is available free with registration.<br /><br /> This report includes <strong>16</strong> pages of action-oriented analysis. What you'll find: <ul class="normalUL"><li>Insight on software that turns servers into storage appliances</li> <li>How to take advantage of SSDs to boost performance without killing budgets</li> </ul> <center><strong><a href="http://reports.informationweek.com/abstract/25/8960/Virtualization/fundamentals-virtualization-lets-you-store-more.html?cid=pub_analyt__iwk_20120827" target="_blank">Get This</a> And <a href="http://reports.informationweek.com/">All Our Reports</a></strong></center> </div> </div> </center></p><br clear="all"> <!-- /KINDLE EXCLUDE --> <P>2012-07-03T16:59:00Zhttp://reports.informationweek.com/abstract/24/8797/Storage-Server/strategy-storage-virtualization-guide.html?cid=RSSfeed_IWK_Authors2011-02-12T00:00:00ZProtecting VMs, and the data that resides on them, needn't hurt performance.http://www.informationweek.com/news/229200218?cid=RSSfeed_IWK_Authors <!-- DIG DEEPER: analytics --> <div style="width:210px; float:right; margin:0; padding:0 0 10px 10px;"> <div style="border:solid 1px #000000; background-color:#ffffff;"> <div style="margin:0; padding:5px; background-color:#CC0000; text-align:center; font-size:1.3em; color:#ffffff; font-weight:bold;">DIG DEEPER</div> <div style="margin:8px 8px 6px 8px; font-weight:bold; font-size:1.2em; text-align:center; color:#cc0000;">Research: 2011 Backup Survey</div> <div style="margin:0 8px 6px 8px; font-weight:bold; text-align:center;">New Possibilities for Data Protection</div> <div style="margin:0 8px 10px 8px; font-weight:bold; text-align:center;"><a href="http://analytics.informationweek.com/abstract/2/5294/Business-Continuity/research-2011-backup-survey.html">Get our <nobr>Analytics Report</nobr></a></div> <div style="margin:0; padding:5px; background-color:#000000; text-align:center; font-size:1em; font-weight:bold; color:#ffffff;"><a href="http://www.informationweekanalytics.com"><span style="color:#ffffff;">See all InformationWeek <nobr>Analytics Reports</nobr></span></a></div> </div> </div> <!-- / DIG DEEPER: analytics --> Server virtualization is a clear win for the data center, but it presents a mixed bag of challenges and opportunities for those charged with backing up their companies' data. For example, in our January InformationWeek Analytics 2011 Backup Technologies Survey, 49% of our 420 respondents say they treat virtual servers the same way they treat their physical servers--installing a backup application agent on each and every virtual machine. </p> <P> Now, that may yield a simple backup architecture, but there's a performance cost when multiple VMs on a virtual server host share network and storage interfaces. When a media server requests backups from several VMs on the same host, all those agents start pulling the requested data concurrently and shooting it out across the network. This will saturate the local and storage area network connections for the virtual server host, not only reducing backup performance but choking other VMs on that host that may be serving users. In addition, this proliferation of agents wastes memory, and when the backup vendor comes out with an updated agent or patch, pity the poor admin. </p> <P> Hypervisors including Microsoft Hyper-V, VMware ESX, and Citrix Xen have host operating systems or service console partitions that enable 28% of our respondents to install backup application agents on the host. While this method does reduce the number of agents that must be installed and maintained, it still requires that the administrator either shut down the virtual machines to back them up or back up the VMs as open files, creating "crash consistent" backups.</p> <P> On the hit list of IT euphemisms, "crash consistent" translates to "only as consistent as the data on disk would be if the server crashed." You'll still lose any data cached in memory.</p> <P> <strong>A Better Way?</strong></p> <P> VMware's first attempt to improve this process, VMware Consolidated Backup (VCB), was a rather clunky piece of software that took a snapshot of the VM and provided access to the snapshot through a Windows proxy server. With vSphere 4, VMware has replaced VCB with a new vStorage backup API that eliminates the need for the Windows server. It can perform image-level or file-by-file backups of VMs by accessing the shared storage that hosts the VMs, taking the load off the virtual server host. The vStorage APIs also allow for block-level incremental backups that are faster and smaller than traditional incremental backups. Just 12% of our survey respondents say they use VMware-specific backup software today, but we expect that number to increase.</p> <P> Meanwhile, three smaller vendors--Vizioncore, now a division of Quest Software; Veeam; and PHD Virtual--have developed backup applications specifically for VMware environments. Their applications leverage the vStorage APIs and add additional functionality, including individual-item restore for applications like Exchange and SharePoint, replication, and simple restores from a series of block incremental backups. Twelve percent of our respondents use these products.</p> <P> <!-- KINDLE EXCLUDE --> <center><img src="http://twimgs.com/informationweek/1291/291F3_BackupStudy_Feb2011_chart5.gif" width="500" height="317" alt="chart: What percentage of your virtual servers are backed up at least weekly?" hspace="0" vspace="0" border="0" /></center></p> <!-- /KINDLE EXCLUDE --> <P> <!-- KINDLE EXCLUDE --> <center> <div style="margin:0; border:solid 1px #cc0000; background-color:#e1e1e1; width:300px; height:100px;"><div style="float:left; padding:10px;"><a href="http://analytics.informationweek.com/issue/1883/informationweek-full-issue-february-14-2011.html"><img src="http://twimgs.com/infoweek/1291/smallcov.jpg" alt="InformationWeek: Feb. 14, 2011 Issue" title="InformationWeek: Feb. 14, 2011 Issue" width="65" height="88" hspace="0" vspace="0" border="0" /></a> </div><div style="margin-top:19px;"><strong> <a href="http://analytics.informationweek.com/issue/1883/informationweek-full-issue-february-14-2011.html">Download a free PDF of <nobr><em>InformationWeek</em> magazine</nobr></a><br /> (registration required)</strong> </div> </div> </center> <!-- /KINDLE EXCLUDE -->2010-08-28T00:00:00ZThink you can't afford BC/DR to rival enterprise-class systems? If you have x86 virtualization installed, you might be surprised.http://www.informationweek.com/news/227100708?cid=RSSfeed_IWK_Authors<!-- KINDLE EXCLUDE --> <!-- Aug. 30. 2010 InformationWeek SMB Green Promo --> <div style="margin:0; padding:0; border-top:dotted 2px #56a643;"> <a href="http://www.informationweek.com/gogreen/083010B/index.jhtml?k=axxe&cid=article_axxe_os"><img src="http://twimgs.com/infoweek/supplement/004/smallcov.jpg" alt="InformationWeek Green - Aug. 30. 2010" width="65" height="87" hspace="0" vspace="0" border="0" align="left" style="margin:12px 33px 8px 15px;" /></a> <a href="http://www.informationweek.com/gogreen/083010B/index.jhtml?k=axxe&cid=article_axxe_os"><img src="http://twimgs.com/infoweek/graphics_library/misc/Green_leaf_88x88.jpg" alt="InformationWeek Green" width="88" height="88" hspace="0" vspace="0" border="0" align="right" style="margin:8px 10px 8px 10px;" /></a> <div style="margin:10px 0 0 0; font-size:1.1em;" align="center"> <strong><a href="http://www.informationweek.com/gogreen/083010B/index.jhtml?k=axxe&cid=article_axxe_os">Download the entire Aug. 30. 2010 issue of <em>SMB</em></a></strong>, distributed in an all-digital format as part of our <a href="http://www.informationweek.com/green/">Green Initiative</a><br /> (Registration required.)<br /> <div style="margin:6px 0 0 0; color:#56a643; font-weight:bold; font-size:1em;">We will plant a tree<br />for each of the first 5,000 downloads.</div> </div> </div> <div style="clear:both; margin:0; padding:0 0 0 0; border-bottom:dotted 2px #56a643;"></div> <!-- / Aug. 30. 2010 InformationWeek SMB Green Promo --> <br /><!-- leave as a br to not interfere w/ the insights boxes --> <!-- /KINDLE EXCLUDE --> <P> <P> <P> What's your best-case scenario for getting back to normal after a worst-case disaster? We first polled small and midsize businesses on that subject back in January 2008; when we revisited our survey, in May, we found there's been some improvement. In 2008, 23% could get mission-critical apps back up in four hours or less. Today, it's up to 33%, based on our <i>InformationWeek Analytics</i> survey of nearly 400 business technology professionals from companies with 1,000 or fewer employees. </p> <P> Other key changes: In 2010, 62% have business continuity/disaster recovery systems in place compared with 55% in 2008. Consolidation has increased; today, 52% are completely centralized, with one main HQ and no branch sites, compared with 44% in 2008. And the number of businesses backing up to tapes that are taken off site dropped a full 16 points, from 63% in 2008 to 47% in 2010. Use of online backup services posted the single biggest gain, up 10 points.</p> <P> One head-scratcher: The number of survey respondents who say their organizations are accountable to one or more government or industry regulations fell in every area, sometimes dramatically. Given the state-level laws that have come on the books since 2008, this is wishful thinking on a massive scale, even for small businesses.</p> <P> Putting a formal business continuity/disaster recovery plan in place and testing it properly costs money, and that's tough to come by nowadays. So to what do we owe improvement in BC/DR? The introduction of new technologies, notably cloud-based storage services, and the maturation of others, like server virtualization and data deduplication, have made effective disaster recovery accessible to a wider swath of businesses than ever before.</p> <P> Widespread use of x86 server vitalization has had the most beneficial effect on the disaster recovery process. An obvious impact is a reduction in the number of physical servers that have to be provisioned, powered, and maintained at a DR location. A few years ago, even the smallest site would have had a dedicated server for each application that needed to be recovered quickly. But now, a single virtual server host can handle multiple applications. It's not only that SMBs can save money on hardware. The reduced size, power, and cooling footprint of a small blade chassis running several virtual server hosts means that branch offices and co-location centers become potential DR sites. That's especially important for small businesses; when we asked respondents to describe their DR setups, the No. 1 answer (with 28%) was another data center or office within the organization. Just 7% use a specialized co-location provider, such as SunGard--down from 14% in 2008.</p> <P> <!-- KINDLE EXCLUDE --> <center><strong>To read the rest of the article,<br /><a href="http://www.informationweek.com/gogreen/083010B/index.jhtml?k=axxe&cid=article_axxe_os">Download the August 30, 2010 <em>InformationWeek</em> SMB digital supplement </a> </strong></center><br clear="all" /></p> <!-- /KINDLE EXCLUDE --> <P> <P> <P> <!-- KINDLE EXCLUDE --> <br clear="all"><center> <div style="margin:0; padding:8px; border:solid 1px #cc0000; width:460px; text-align:left;"> <div style="margin:0; padding:5px; background-color:#CC0000; text-align:center; font-size:1.3em; color:#ffffff; font-weight:bold;"><a href="http://analytics.informationweek.com/abstract/2/3956/Business-Continuity/research-bc-dr-for-smbs.html" target="_blank" style="color:#ffffff;">Research: BC/DR for SMBs</a></div> <div style="margin:8px;"> <center><strong><a href="informationweek .com/analytics/smbbackups">Small Favors: Technology Advances SMBs' BC/DR Plans</a></strong> <br /><img src="http://twimgs.com/informationweek/supplement/004/004_SMB_DCDRreportcover_110.jpg" width="110" height="110" hspace="0" vspace="59" border="0" align="right" style="margin:8px 0 9px 9px;" /><br /> <strong><a href="http://analytics.informationweek.com/abstract/2/3956/Business-Continuity/research-bc-dr-for-smbs.html">Subscribe</a></strong> and get our full report on SMB BC/DR. This report includes <strong>40</strong> pages of action-oriented analysis, packed with <strong>24</strong> charts. <br /><br /></center> What you'll find:<br /> <ul><li>A game plan for prioritizing your data protection efforts.</li> <li>An in-depth discussion of DR site options</li> <li>Synchronous, asynchronous, or snapshot? We help you decide</li> </ul></strong> <center><strong><a href="http://analytics.informationweek.com/abstract/2/3956/Business-Continuity/research-bc-dr-for-smbs.html">Get This</a> And <a href="http://analytics.informationweek.com/">All Our Reports</a></strong><br /></center> </div> </center></div> </p> <!-- /KINDLE EXCLUDE -->2010-08-28T00:00:00ZThink you can't afford BC/DR to rival enterprise-class systems? If you have x86 virtualization installed, you might be surprised.http://www.informationweek.com/news/227100691?cid=RSSfeed_IWK_AuthorsWhat's your best-case scenario for getting back to normal after a worst-case disaster? We first polled small and midsize businesses on that subject back in January 2008; when we revisited our survey, in May, we found there's been some improvement. In 2008, 23% could get mission-critical apps back up in four hours or less. Today, it's up to 33%, based on our <i>InformationWeek Analytics</i> survey of nearly 400 business technology professionals from companies with 1,000 or fewer employees. </p> <P> Other key changes: In 2010, 62% have business continuity/disaster recovery systems in place compared with 55% in 2008. Consolidation has increased; today, 52% are completely centralized, with one main HQ and no branch sites, compared with 44% in 2008. And the number of businesses backing up to tapes that are taken off site dropped a full 16 points, from 63% in 2008 to 47% in 2010. Use of online backup services posted the single biggest gain, up 10 points.</p> <P> One head-scratcher: The number of survey respondents who say their organizations are accountable to one or more government or industry regulations fell in every area, sometimes dramatically. Given the state-level laws that have come on the books since 2008, this is wishful thinking on a massive scale, even for small businesses.</p> <P> Putting a formal business continuity/disaster recovery plan in place and testing it properly costs money, and that's tough to come by nowadays. So to what do we owe improvement in BC/DR? The introduction of new technologies, notably cloud-based storage services, and the maturation of others, like server virtualization and data deduplication, have made effective disaster recovery accessible to a wider swath of businesses than ever before.</p> <P> Widespread use of x86 server vitalization has had the most beneficial effect on the disaster recovery process. An obvious impact is a reduction in the number of physical servers that have to be provisioned, powered, and maintained at a DR location. A few years ago, even the smallest site would have had a dedicated server for each application that needed to be recovered quickly. But now, a single virtual server host can handle multiple applications. It's not only that SMBs can save money on hardware. The reduced size, power, and cooling footprint of a small blade chassis running several virtual server hosts means that branch offices and co-location centers become potential DR sites. That's especially important for small businesses; when we asked respondents to describe their DR setups, the No. 1 answer (with 28%) was another data center or office within the organization. Just 7% use a specialized co-location provider, such as SunGard--down from 14% in 2008.</p> <P> Companies that virtualize their production servers can create a replication pair for each set of host servers, rather than for each application. And administrators will have no problem restoring servers to dissimilar hardware using virtual server and data migration tools such as vMotion and XenMotion. They also allow IT to quickly add servers to the cluster and rebalance the load, so organizations can bring systems back online faster and more gracefully. </p> <P> We have this to say to the 41% of businesses that don't plan to leverage virtualization in their BC/DR strategies: Reconsider. That's especially true if, like many SMBs, your storage team is primarily responsible for BC/DR. When companies task storage admins with disaster recovery, the lion's share of planning resources are focused on preserving the organization's data, not on ensuring applications can be brought back into operation quickly.</p> <P> Even if the system drive of a physical server has been replicated, bringing that server back online is a time-consuming task. If a similar device is at the DR site, it will have to be connected to the replicated system drive and booted. An administrator must then log in and reset the network configuration, since the DR site is on a different subnet and the recovery server has a different MAC address than the original device. For applications like Microsoft Exchange that are tightly tied to other network functions, including Active Directory, the process can be even more challenging. </p> <P> Other new technologies, such as data deduplication and WAN optimization, can greatly reduce the bandwidth needed to synchronize data stores between sites. The actual reduction in required bandwidth is highly dependent on the data being replicated and can range from as little as 5% to as much as 95%, so testing is required. Vendors such as Cisco, Data Domain, Hewlett-Packard, Riverbed, and others offer a range of products that help conserve WAN bandwidth.</p> <P> SMBs use a variety of backup technologies, with the most popular being off-site backup to tape (47%) followed by on-site backup to tape (34%). While use of both methods has dropped significantly, tape is still the primary media for backing up critical data, such as e-mail archives, which are vital for documentation and compliance. Some SMBs are embracing cloud-based online backup services, others are skeptical. Us? We're big believers in cloud backup and storage for SMBs. Companies with limited IT resources don't handle tape well and should move to disk and online ASAP. If you're still taking tapes off site weekly, your best-case recovery point from a disaster might be last Wednesday. And that's not nearly good enough. </p> <P>2009-11-28T00:00:00ZOptions abound for this new technology. We help you sort them out.http://www.informationweek.com/news/221901163?cid=RSSfeed_IWK_AuthorsMention cloud storage to most IT professionals and they think of Internet services like Amazon S3 and Nirvanix that store your data in their data centers.</p> <P> But a storage cloud doesn't have to be public. A wide range of private cloud storage products have been introduced by vendors, including name-brand companies such as EMC, with its Atmos line, and smaller players like ParaScale and Bycast. Other vendors are slapping the "cloud" label on existing product lines. Given the amorphous definitions surrounding all things cloud, that label may or may not be accurate. What's more important than semantics, however, is finding the right architecture to suit your storage needs.</p> <P> <P> <CENTER> <A HREF="http://analytics.informationweek.com/abstract/5/1693/Cloud-Computing/private-cloud-storage-decoded.html?cid=IWKRPT" target="_blank"><IMG SRC="http://twimgs.com/infoweek/promo/iwkreports.gif" ALT="InformationWeek Reports" WIDTH="278" HEIGHT="25" BORDER="0"></A></CENTER></P> <P> A prototypical cloud storage system is made up of a number of x86 servers, each with its own storage, most commonly using four to 16 SATA drives. Users and their applications access the system through standard file access protocols like CIFS and NFS or via object storage and retrieval protocols like SOAP and REST.</p> <P> The storage nodes in a private cloud are linked together with a layer of smart software, which performs several functions. First, it maintains a global name space that allows all the storage in the cluster to be accessed as a single entity, so that administrators can add storage capacity on the back end without having to tell applications at the front end how to reach it. The software also handles drive failures and keeps data available to applications and end users.</p> <P> A private cloud storage infrastructure should also be able to scale from hundreds of terabytes to multiple petabytes. That level of scalability is achieved not with a forklift upgrade, but simply by adding more servers as they're needed.</p> <P> This architecture provides two major benefits. First, storage administrators can configure and provision new storage nodes quickly and inexpensively. Second, administrators can add capacity only as demand requires, instead of purchasing additional disk space to meet anticipated future growth and then having that capacity sit idle in the present.</p> <P> However, there are also trade-offs. Cloud storage is best suited to unstructured data, such as medical images, engineering drawings, and Office documents. For another, because each x86 server isn't as reliable as a high-end enterprise disk array, a private cloud must store copies of the data on multiple nodes. This requires more raw disk space than an enterprise disk array using a RAID-5 or 6 system. For example, if you set a policy for your private cloud to keep three copies of a 60-GB file for data protection, it would require 180 GB of disk, whereas a 6+2 RAID-6 system would need just 80 GB.</p> <P> <strong>Beyond Low Cost</strong></p> <P> <!-- Our Take --> <div style="margin:0; padding:0 0 10px 10px; width:260px; float:right;"> <div style="border:1px solid #000000; padding:0;"> <div style="margin:0; padding:4px; font-size:1.4em; text-align:center; color:#ffffff; background-color:#000000;"> <strong>Our Take</strong> </div> <div style="margin:0; padding:4px; font-size:1.2em; text-align:center; color:#ffffff; background-color:#CC0000;"> <strong>PRIVATE CLOUD STORAGE</strong> </div> <div style="margin:0; padding:8px;"> Can reduce up-front hardware and administrative costs <div style="margin:10px 0 10px 0; padding 0; border-bottom:solid 1px #666666;"></div> Takes advantage of low-cost servers and storage <div style="margin:10px 0 10px 0; padding 0; border-bottom:solid 1px #666666;"></div> Makes it easier to add storage capacity <div style="margin:10px 0 10px 0; padding 0; border-bottom:solid 1px #666666;"></div> Most appropriate for large volumes of unstructured data </div> </div> </div> <!-- / Our Take --> Private cloud software vendors are focused on finding ways to differentiate themselves from their competitors. For instance, Cleversafe says it gets around the RAID issue through unique dispersal algorithms that ensure data availability with less than 40% overhead.</p> <P> Several other vendors include location-aware policy engines that copy data to nodes in specific geographical locations. Data Direct Networks' Web Object Store, Bycast's StorageGrid, and EMC's Atmos systems can specify that two copies of each object in a folder should be stored in New York and Los Angeles, and that copies also should be stored in two other locations.</p> <P> This not only protects data from data center failures but can also put objects on storage clusters close to the users who need them. Bycast's policy engine takes this notion one step further by including elements, such as storage tiering, that can migrate objects from more-expensive to less-expensive disk, and even to and from tape.</p> <P> Organizations planning to offer private cloud storage services to internal departments may want to consider multitenant features that allow storage to be partitioned among different groups. For example, IT could carve out one section of the private cloud for HR and another for marketing, and then charge those departments based on usage. This means having delegated administration models and/or virtual servers that restrict each group's access and visibility to only their own data and the resources assigned to them. A multitenant storage system should also include accounting features that collect usage data, such as peak utilization, that will help IT in determining chargebacks.</p><strong>Partly Cloudy</strong></p> <P> Given the attention that cloud computing garners these days, some vendors are rebranding existing offerings as private cloud options. This can be frustrating for potential buyers, but religious arguments over what constitutes a cloud are less important than features, capabilities, and cost.</p> <P> Caringo and HDS have repositioned their content addressable storage (CAS) and redundant array of independent nodes (RAIN) systems as private cloud storage. There are some similarities. For instance, CAS/RAIN architectures tend to be built with less-expensive disks than you'd find in an enterprise SAN.</p> <P> However, vendors have traditionally positioned CAS/RAIN architectures for archiving and compliance. Those use cases require more-advanced features than most private cloud providers offer, such as deduplication, or the ability to set retention and disposition policies or use hash algorithms to demonstrate that objects haven't been changed after they're saved. These advanced features let vendors charge a premium, which starts to push these products outside the low-cost boundary of a private cloud. In addition, the amounts of data CAS/RAIN storage systems are intended to hold are usually smaller, and have lower performance requirements, than a private cloud architecture.</p> <P> <!-- The Essentials --> <div style="margin:0; padding:0 0 10px 10px; width:250px; float:right;"> <div style="border:1px solid #000000; padding:0; font-size:.8em;"> <div style="margin:0; padding:4px; font-size:1.4em; text-align:center; color:#ffffff; background-color:#000000;"> <strong>PRIVATE CLOUD<br />STORAGE OPTIONS</strong> </div> <div style="margin:0; padding:6px; background-color:#c2d9ef;"> <span style="color:#CC0000; font-weight:bold;">Bycast StorageGrid</span><br /> Software; location aware; supports multitenancy and multiple data tiers <div style="margin:4px 0 4px 0; padding 0; border-bottom:dashed 1px #000000;"></div> <span style="color:#CC0000; font-weight:bold;">Caringo CAStor</span><br /> CAS/RAIN software; replicates among clusters; optional CIFS/NFS gateway software <div style="margin:4px 0 4px 0; padding 0; border-bottom:dashed 1px #000000;"></div> <span style="color:#CC0000; font-weight:bold;">Cleversafe</span><br /> Software; disperses data slices across multiple locations; iSCSI interface <div style="margin:4px 0 4px 0; padding 0; border-bottom:dashed 1px #000000;"></div> <span style="color:#CC0000; font-weight:bold;">Data Direct Networks Web Object Scaler</span><br /> Appliance-based object store; location-aware policy engine; up to 60 TB per node <div style="margin:4px 0 4px 0; padding 0; border-bottom:dashed 1px #000000;"></div> <span style="color:#CC0000; font-weight:bold;">EMC Atmos</span><br /> Integrated hardware/ software; distributed objects; policy engine; supports multitenancy; minimum config is 120 TB <div style="margin:4px 0 4px 0; padding 0; border-bottom:dashed 1px #000000;"></div> <span style="color:#CC0000; font-weight:bold;">HDS Hitachi Content Platform</span><br /> CAS/RAIN appliance; internal or external storage; replicates between clusters; supports multitenancy <div style="margin:4px 0 4px 0; padding 0; border-bottom:dashed 1px #000000;"></div> <span style="color:#CC0000; font-weight:bold;">IBM Business Storage Cloud</span><br /> Cluster file system with integrated product and services; can use IBM XIV grid back end <div style="margin:4px 0 4px 0; padding 0; border-bottom:dashed 1px #000000;"></div> <span style="color:#CC0000; font-weight:bold;">ParaScale Cloud Storage Software</span><br /> Software; distributed object copies; replicates among clusters <div style="margin:4px 0 4px 0; padding 0; border-bottom:dashed 1px #000000;"></div> <span style="color:#CC0000; font-weight:bold;">Symantec FileStore</span><br /> Cluster file system software; uses shared storage; replicates file systems; supports multiple tiers </div> </div> </div> <!-- / The Essentials --> The CAS/RAIN vendors aren't the only ones using cloud labels to fog up product categories. Vendors like IBM and Symantec have repackaged their clustered file systems into private clouds. Symantec FileStore software wraps Storage Foundation, and its integrated VxFS clustered file system, in a package that's easier to install and manage. IBM's Smart Business Storage Cloud leverages its GPFS clustered file system along with XIV clustered block storage (and of course, IBM services).</p> <P> While cluster file systems can deliver impressive performance, their reliance on expensive back-end storage makes them relatively pricey compared with RAIN architectures. Cluster file systems are more appropriate to applications, like render farms, that require high performance for individual clients.</p> <P> <strong>Pick A Package</strong></p> <P> Organizations that want to get private cloud storage off the ground quickly, or prefer the comfort of one throat to choke, should consider integrated systems like Hitachi's Content Platform, EMC's Atmos, or Data Direct Networks' Web Object Store. These products come complete with storage hardware, software, processors--and in the case of Atmos, even the rack.</p> <P> Those looking for cloud economics may prefer software like Bycast's StorageGrid, ParaScale's Storage Cloud, or Caringo's CAStor. Because these vendors charge for their software on a per- gigabyte basis, users can easily match capacity to cost. Meanwhile, Cleversafe sells pre-configured access, storage, and management nodes, and the adventurous can use the open source community version from Cleversafe.org.</p> <P> Private cloud storage systems can bring cloud economics to the data center, allowing corporate IT to retain control over data, security, and reliability. These new architectures promise to not only reduce the up-front cost of storing many terabytes of unstructured data but also reduce the amount of manpower required to manage it.</p> <P> <strong>Howard Marks is chief scientist at Networks Are Our Lives, a consulting firm.</strong></p>2009-09-05T00:00:11ZFlash-based solid-state drives are fast and green -- and expensive. Get best results with a data classification strategy.http://www.informationweek.com/news/219501231?cid=RSSfeed_IWK_AuthorsBecause the storage subsystem is the performance bottleneck for most commercial applications, system designers have sought to speed the flow of data from disk to main memory using a variety of schemes. The latest move relies on solid-state disk technology. Over the past year, flash-memory-based solid-state devices have emerged that deliver data in a few microseconds; are significantly less expensive than RAM on a per-gigabyte basis; and, like disks, store data reliably when powered off.</p> <P> Enterprise-class flash devices are still pricey--averaging more than $100 per gigabyte. But for those with deep pockets and a real need for speed, these systems can deliver up to 45,000 read I/O operations per second, or 16,000 write IOPS, compared with the 170 IOPS typical of a 15-000 RPM drive. A single mirrored pair of SSDs can outperform 100 spinning disks that would cost several times as much after drive enclosures, array features, software that's licensed by capacity, and other so-called slot costs are figured in. Flash devices also can help you go green because that pair of high-end SSDs will use much less power, and generate less heat that the data center cooling system must remove, compared with a group of 15,000-RPM drives delivering the same IOPS.</p> <P> <P> <CENTER> <A HREF="http://informationweekreports.com/shared/download.jhtml?id=178300032&cat=iwkr_servers&doc_id=219501231?cid=IWKRPT" target="_blank"><IMG SRC="http://twimgs.com/infoweek/promo/iwkreports.gif" ALT="InformationWeek Reports" WIDTH="278" HEIGHT="25" BORDER="0"></A></CENTER></P> <P> Vendors from EMC to Xiotech are debuting innovative SSD technologies suitable for enterprise data centers and branch offices. But there are a few downsides. Long-term reliability is unproven, the bottom-line cost on a per-gigabyte basis could send your CFO into cardiac arrest, and advances in data classification and tiering are still needed to gain maximum benefit. IT groups need to be aware of the trade-offs.</p> <P> <strong>How Flash Works</strong><br /> While flash is semiconductor memory, like the RAM that makes up a computer's main memory, it doesn't allow direct read and write access to each byte, the way RAM does. Just as a disk drive is divided into sectors, the NAND flash chips typically used in SSDs are organized into pages, typically of 4 KB each. These pages are in turn collected into blocks of 256 KB to 1 MB.</p> <P> Data can be read from the flash memory on a page-by-page basis and written to empty pages. However, to overwrite data in a previously used page, the entire block containing that page must be erased--a relatively slow process. If other pages in that block contain valuable data, that information must either be relocated to pages in another block or loaded into RAM cache in the SSD and written back once the block has been erased. This results in flash-memory devices being three to 10 times slower to write data than to read it. In addition, the high voltages needed to erase blocks cause wear on the drives' microscopic transistors and connections, eventually wearing them out.</p> <P> <!-- InformationWeek reports refer --> <div style="width:210px; float:right; margin:0; padding:0 0 10px 10px;"> <div style="border:solid 1px #000000; background-color:#ffffff;"> <div style="margin:0; padding:5px; background-color:#2a45a9; text-align:center; font-size:1.3em; color:#ffffff; font-weight:bold;">DIG DEEPER</div> <div style="margin:8px 8px 6px 8px; font-weight:bold; font-size:1.2em; text-align:center; color:#cc0000;">Put Data On A Diet</div> <div style="margin:0 8px 8px 8px; font-weight:bold; text-align:center;">Find out how deduplication can trim bloated file systems.</nobr></div> <div style="margin:0 8px 10px 8px; font-weight:bold; text-align:center;"><a href="http://informationweekreports.com/shared/download.jhtml?id=175300037&cat=iwkr_servers&doc_id=216500085">Download this<br />InformationWeek Report</a></div> <div style="margin:0; padding:5px; background-color:#000000; text-align:center; font-size:1em; font-weight:bold; color:#ffffff;"><a href="http://www.informationweekreports.com"><span style="color:#ffffff;">See all our <nobr>InformationWeek Reports</nobr></span></a></div> </div> </div> <!-- / InformationWeek reports refer --> Flash memory comes in two basic varieties, differentiated mainly by the ability to handle multiple erase cycles. As the name implies, single-level cell, or SLC, flash stores 1 bit in each memory cell, while multilevel cell memory stores 2 or 3 bits of data in each cell.</p> <P> Storing 2 or 3 bits per cell increases density, and therefore reduces cost, but it also slows access and reduces the longevity of the device. Therefore, most server and array flash systems use SLC technology.</p> <P> Even among SSDs using the same flash technologies--or even the same flash chips--performance can vary significantly, however. A main variable is that individual vendors design the controllers that put a storage interface on their flash chips.</p> <P> <table border="0" cellpadding="4" cellspacing="0" width="100%"> <tr bgcolor="#ffffff""> <td colspan="6"> <div style="margin:0; border-bottom:1px solid black; color:#cc0000; text-transform:uppercase; font-size:1.3 em;"><strong>Priced For Speed</strong></div> </td> </tr> <tr bgcolor="#ffffff" valign="bottom" align="center"> <td>&nbsp;</td> <td><span style="color:#cc0000; font-weight:bold;">Capacity</span></td> <td><span style="color:#cc0000; font-weight:bold;">Cost</span></td> <td><span style="color:#cc0000; font-weight:bold;">Read<br />IOPS</span></td> <td><span style="color:#cc0000; font-weight:bold;">Write<br />IOPS</span></td> <td><span style="color:#cc0000; font-weight:bold;">Active<br />Power</span></td> </tr> <tr bgcolor="#ccccff" valign="top" align="center"> <td align="left"><strong>STEC Zeus IOPS</strong></td> <td>146 GB</td> <td>$16,000</td> <td>45,000</td> <td>16,000</td> <td>8.4 W</td> </tr> <tr bgcolor="#ffffff" valign="top" align="center"> <td align="left"><strong>Intel X-25E</strong></td> <td>64 GB</td> <td>$699</td> <td> 35,000</td> <td>3300</td> <td>2.6 W</td> </tr> <tr bgcolor="#ccccff" valign="top" align="center"> <td align="left"><strong>Seagate Cheetah 15K</strong></td> <td>146 GB</td> <td>$350 </td> <td> 185</td> <td> 169</td> <td>17 W</td> </tr> <tr bgcolor="#ffffff" valign="top" align="center"> <td align="left"><strong>TMS RamSan-20</strong></td> <td>450 GB</td> <td>$15,200</td> <td>120,000</td> <td>50,000</td> <td>15 W</td> </tr> <tr bgcolor="#ccccff" valign="top" align="center"> <td align="left"><strong>Fusion-io IoDrive</strong></td> <td>160 GB</td> <td>$7,200</td> <td>116,000</td> <td>83,174</td> <td>19 W</td> </tr> </table></p><strong>Vendors Flock To Flash</strong><br /> EMC started the trend of SSD adoption by adding STEC's Zeus IOPS SSD to its Symmetrix, Clariion, and Celerra product lines about 15 months ago. Following EMC's lead, most array vendors, including IBM, HDS, and Hewlett-Packard, are replacing standard Fibre Channel drives in their arrays with SSDs.</p> <P> The benefit for storage administrators is that they can build logical unit number, or LUN, RAID drives from these flash drives and move their most I/O-intensive data to these new, blazingly fast LUNs.</p> <P> Of course, identifying this I/O-intensive data and moving it to flash isn't always a simple matter. You'll need someone with detailed knowledge of the organization's data and a database administrator who can move Oracle tables, or high-use portions of tables, to the new flash LUN. Users of applications like Exchange that treat the entire database as a single file will have to transfer the whole enchilada to gain benefits.</p> <P> The next step in accommodating SSDs is for vendors to adopt automatic storage tiering in their devices. Then, the storage system will track what data needs faster access and automatically move it to faster SSD LUNs, lightening the load on admins. There is movement here: Symantec's VxFS file system, a component of its Storage Foundation storage management package, can move frequently accessed files to faster disk, while Compellent's Storage Center has supported automated tiering for block data for over a year. EMC has announced its Fully Automated Storage Tiering technology, which will move files on Celerra devices later this year and manage block data on Clariion and Symmetrix boxes in 2010.</p> <P> Sun's (now Oracle's) Open Storage products currently employ high-performance SLC SSDs to hold the frequently accessed file system logs for its ZFS file system and can use hundreds of gigabytes of lower-cost multilevel cell memory, which is slow to write but fast to read, as a read cache. This combination allows Open Storage NAS systems with flash and SATA drives to perform like rival devices with costlier 10,000- or 15,000-RPM drives.</p> <P> Bucking the trend toward using a small number of very fast but expensive SSDs, Pillar Data Systems and Dell/EqualLogic have chosen to implement whole enclosures of lower-cost, but still SLC-based, SATA SSDs from Intel and Samsung. While these lower-cost drives deliver just one-sixth the write speed of STEC's, they're a small fraction of the cost--$10 per gigabyte for Intel X25-E vs. $110 per gigabyte for STEC. They can deliver plenty of performance at a price that branch-office and midsize customers can afford.</p> <P> Rather than bundling flash chips into modules that emulate disk drives in size and interface, another group of vendors, led by Fusion-io and RAM SSD pioneer Texas Memory Systems, are putting flash memory on PCI Express cards. These products can achieve astounding performance in both throughput and latency by bypassing drive interface electronics, RAID controllers, and SAN interconnects.</p> <P> <!-- Our Take --> <div style="margin:0; padding:0 0 10px 10px; width:260px; float:right;"> <div style="border:1px solid #000000; padding:0;"> <div style="margin:0; padding:4px; font-size:1.4em; text-align:center; color:#ffffff; background-color:#000000;"> <strong>Our Take</strong> </div> <div style="margin:0; padding:4px; font-size:1.2em; text-align:center; color:#ffffff; background-color:#CC0000;"> <strong>SOLID-STATE DRIVES</strong> </div> <div style="margin:0; padding:8px; text-align:center;"> SSDs are performance monsters--a single mirrored pair can outperform 100 <nobr>spinning disks.</nobr> <div style="margin:10px 0 10px 0; padding 0; border-bottom:solid 1px #666666;"></div> They provide green benefits such as reduced power and <nobr>cooling costs.</nobr> <div style="margin:10px 0 10px 0; padding 0; border-bottom:solid 1px #666666;"></div> Major vendors are bringing more SSDs to market, driving <nobr>down prices.</nobr> <div style="margin:10px 0 10px 0; padding 0; border-bottom:solid 1px #666666;"></div> That said, they're still expensive. SSDs will take a bite out of your <nobr>storage budget.</nobr> <div style="margin:10px 0 10px 0; padding 0; border-bottom:solid 1px #666666;"></div> Long-term viability for SSDs is <nobr>still unproven.</nobr> </div> </div> </div> <!-- / Our Take --> While PCIe flash drives are fast, they're also direct-attached storage that's owned by a single server. This is at odds with the move toward ubiquitous server virtualization, which relies on a shared storage back end--either SAN or NFS--something current PCIe flash systems can't support.</p> <P> <strong>Time To Flash?</strong><br /> We're just in the first generation of data center flash implementations, and there's no consensus on how best to take advantage of this new technology. Even so, flash SSDs are too powerful a tool for enterprise data center engineers to ignore, and the devices based on these drives will only get more compelling as prices fall and storage system designers integrate their advantages more tightly into next-generation products.</p> <P> For now, IT groups that can easily identify, and relocate, 5% or more of their stored data that require significantly higher I/O rates should be looking seriously at adding a flash-based Tier 0 to their storage infrastructures. Companies that can't yet easily identify their hot spots should launch a data classification project to do so, while closely following the development of automated tiering.</p> <P> <em><strong>Howard Marks is chief scientist at <a href="http://www.deepstorage.net" targrt="_blank">DeepStorage.net</a>, a testing lab and analyst firm.</strong></em></p>2009-06-06T00:00:12ZNew offerings and approaches feed multicore servers the virtual bandwidth they need.http://www.informationweek.com/news/217701835?cid=RSSfeed_IWK_AuthorsMulticore servers allow organizations to consolidate 20 or more production servers into each virtual server host, saving space, power, and staff resources in the process. But feeding these servers enough I/O bandwidth has become a major headache for many data centers.</p> <P> Some experts recommend Gigabit Ethernet connections for each processor core on the host to ensure sufficient bandwidth, dedicated connections for management and virtual machine migration, plus additional Fibre Channel or Gigabit Ethernet connections for storage, resulting in cable and switch port sprawl.</p> <P> <P> <CENTER> <A HREF="http://informationweekreports.com/shared/download.jhtml?id=176700012&cat=iwkr_virtualization&doc_id=217701835?cid=IWKRPT" target="_blank"><IMG SRC="http://twimgs.com/infoweek/promo/iwkreports.gif" ALT="InformationWeek Reports" WIDTH="278" HEIGHT="25" BORDER="0"></A></CENTER></P> <P> Leading vendors, including many Fibre Channel providers, are pushing enhanced 10-Gbps Ethernet with Fibre Channel over Ethernet (FCoE) as the solution. Granted, 10-Gbps Ethernet addresses the bandwidth issue, and FCoE converged network adapters can handle the storage side of the equation. However, the standards for both FCoE and Date Center Bridging, the IEEE name for the Ethernet enhancements FCoE requires, haven't yet been ratified.</p> <P> <!-- reports promo right --> <div style="width:210px; float:right; margin:0; padding:0 0 10px 10px;"> <div style="border:solid 1px #000000; background-color:#ffffff;"> <div style="margin:0; padding:5px; background-color:#2a45a9; text-align:center; font-size:1.3em; color:#ffffff; font-weight:bold;">DIG DEEPER</div> <div style="margin:8px 8px 6px 8px; font-weight:bold; font-size:1.2em; text-align:center; color:#cc0000;">Virtualization Showdown</div> <div style="margin:0 8px 8px 8px; font-weight:bold; text-align:center;">We put leading VM platforms to the test.</div> <div style="margin:0 8px 10px 8px; font-weight:bold; text-align:center;"><a href="http://informationweekreports.com/shared/download.jhtml?id=176000008&cat=iwkr_virtualization&doc_id=InformationWeek_Analytics_Alert_ServerVM">Download the report to <nobr>see the results</nobr></a></div> <div style="margin:0; padding:5px; background-color:#000000; text-align:center; font-size:1em; font-weight:bold; color:#ffffff;"><a href="http://www.informationweekreports.com"><span style="color:#ffffff;">See all our<br /><em>InformationWeek</em> Reports</span></a></div> </div> </div> <!-- / reports promo right --> Other vendors have proposed solutions to the virtual server I/O problem that could be more cost-effective and flexible. These systems create multiple virtual-aware adapters (virtual network interface cards and virtual host bus adapters) assigned to virtual machines. As the VMs migrate from host to host, the virtual NICs and virtual HBAs let them keep their MAC address or World Wide Name, alleviating some network security and zoning issues of VMs.</p> <P> Virtualization-aware I/O devices' support for quality of service means that virtual NICs and HBAs supporting critical or latency-sensitive applications can use reserved bandwidth and take higher priority than users surfing Facebook while sharing a single physical connection.<br clear="all" /></p> <P><strong>To InfiniBand And Beyond</strong><br /> The first alternative to FCoE, championed by InfiniBand vendors Voltaire and Mellanox, connects the servers to InfiniBand switches and uses InfiniBand-to-Ethernet and Fibre Channel bridges to connect to a data center's LAN and SAN resources.</p> <P> InfiniBand is best known as the high-speed (up to 40 Gbps), lossless, low-latency interconnect at the heart of most high-performance computing clusters. Advocates promote it as the Swiss Army knife of networks, good for remote direct memory access (RDMA) in high-performance computing, IP transport, and even storage connections in applications. On the plus side, InfiniBand solutions bring two to four times the bandwidth of even 10-Gbps Ethernet for server-to-server I/O functions like VM migration.</p> Voltaire has a 10-Gbps Ethernet/InfiniBand line card that has two 10-Gbps Ethernet ports and 22 10/20-Gbps InfiniBand ports, for its Grid Director 2004 and 2012 switches. The host can use an IP-over-InfiniBand driver and the InfiniBand switch provides Layer 2 bridging and Layer 3 and 4 routing to the Ethernet ports for data networking. For storage access, Voltaire has a storage router with two 10/20 Gbps InfiniBand ports and four 4-Gbps Fibre Channel ports. Hosts use iSCSI or iSER (iSCSI Extensions for RDMA) and the storage router maps iSCSI targets to Fibre Channel logical unit numbers.</p> <P> At first glance, Mellanox's BridgeX looks very much like the Voltaire approach, with a bridge appliance that has four 40-Gbps InfinBand ports facing the servers and 16 8-Gbps Fibre Channel ports or 12 10-Gbps Ethernet ports facing networks and/or storage. Mellanox's ConnectX cards have dual personalities, so each port can act as a 40-Gbps InfiniBand HBA, like the Voltaire solution, or a 10-Gbps Ethernet port.</p> <P> The BridgeX is a stateless system where the Mellanox network and storage driver encapsulate Fibre Channel Protocol (FCP) or Ethernet packets inside InfiniBand packets. Where an iSCSI-based system needs to terminate the IP session, unwrap the IP from the SCSI data and rewrap it in FCP, BridgeX just has to peel off the InfiniBand wrapper.</p> <P> ConnectX is a 10-Gbps card as well, so the server can take advantage of Mellanox's FCoE storage drivers to connect to an FCoE switch like Cisco's Nexus 5000 or Brocade's 8000, or use Mellanox's BridgeX to gateway FCoE traffic to a Fibre Channel switch. When used this way, BridgeX emulates N-Ports to the Fibre Channel switch looking like a series of Fibre Channel HBAs. However, unlike an FCoE switch, BridgeX doesn't provide naming or other fabric services.</p> <P> While Xsigo uses InfiniBand, its I/O Director isn't an InfiniBand-to-Ethernet and Fibre Channel bridge, but instead uses InfiniBand as the connection media to Gigabit and 10-Gbps Ethernet, and to Fibre Channel modules in the I/O Director. Host systems connect to the 24 20-Gbps InfiniBand ports, but rather than iSER or other InfiniBand drivers, they use Ethernet drivers for the Ethernet ports. Because the Fibre Channel modules use QLogic silicon, QLogic's SANsurfer and Fibre Channel drivers also let admins manage virtual HBAs with SAN management tools. The I/O Director has 15 slots for two-port 4-Gbps Fibre Channel, four-port Gigabit, and 10-Gbps Ethernet modules. Organizations that need higher server-to-I/O densities can use InfiniBand switches between servers and the I/O Director to connect hundreds of servers to a single I/O Director.</p> <P> Betting against Ethernet has historically been a good way to lose money -- see Token Ring and ATM LANs, for example -- and all three of the vendors using InfiniBand as part of their products are integrating 10-Gbps Ethernet into their portfolios. Mellanox's road map is pretty clear; Voltaire is leveraging its InfiniBand expertise to build high-port-count, low-latency 10-Gbps switches; and Xsigo has been hinting that a future version of I/O Director will use 10-Gbps Ethernet instead of InfiniBand.</p> <P> <!-- The Essentials --> <table border="0" width="100%" bgcolor="#000000" cellpadding="3" cellspacing="1"> <tr bgcolor="#000000"> <td colspan="3" align="center"><span style="font-size:1.2em; font-weight:bold; color:#ffffff;">The Essentials</span></td> </tr> <tr bgcolor="#CC0000"> <td colspan="3" align="center"><span style="font-size:1.4em; color:#ffffff; text-transform:uppercase;"><strong>I/O Virtualization Technologies </strong></span></td> </tr> <tr bgcolor="#CCCCCC" valign="top" align="center"> <td bgcolor="#666666">&nbsp;</td> <td bgcolor="#CCCCCC"><span style="font-size:1.2em; color:#CC0000; text-transform:uppercase;"><strong>Pros</strong></span></td> <td bgcolor="#CCCCCC"><span style="font-size:1.2em; color:#CC0000; text-transform:uppercase;"><strong>Cons</strong></span></td> </tr> <tr bgcolor="#FFFFFF" align="center"> <td bgcolor="#666666"><span style="color:#ffffff;"><strong>10-Gbps Ethernet with FCoE</strong></span></td> <td> Supplies bandwidth plus storage<br /><br /> Major vendor support </td> <td> Standards aren't final for FCoE or needed extensions<br /><br /> Potential political issues between network and storage groups </td> </tr> <tr bgcolor="#FFFFFF" align="center"> <td bgcolor="#666666"><span style="color:#ffffff;"><strong>InfiniBand with Fibre Channel bridges</strong></span></td> <td> Consolidates network and storage<br /><br /> High server-to-server bandwidth, low latency </td> <td> Servers see InfiniBand drivers, not Ethernet/ Fibre Channel </td> </tr> <tr bgcolor="#FFFFFF" align="center"> <td bgcolor="#666666"><span style="color:#ffffff;"><strong>Switches to external I/O chassis containing additional PCIe slots</strong></span></td> <td> PCIe card lets servers use unmod-ified I/O device drivers<br /><br /> Can support additional I/O, such as video </td> <td> Emerging technology from smaller vendors </td> </tr> </table> </p> <!-- / The Essentials --> </p><strong>Slot Machines</strong><br /> Another group of vendors, lead by NextIO with newcomers Aprius and VirtenSys, promise products that will extend the server's PCIe slots through a switch to an external I/O chassis containing additional PCIe slots. Conceptually much like the PCI SIG's Multiple Root-I/O Virtualization (MR-IOV) standard (see sidebar), but without the need for I/O cards to have MR-IOV support, these systems use a low-cost -- around $200, versus $1,500 for a converged network adapter -- stateless PCIe extender card so servers use the I/O devices drivers unmodified.</p> <P> PCIe-based solutions can share any Single Root-I/O Virtualization (SR-IOV) compatible card allocating their virtual interfaces to hosts as virtual devices; cards that don't support IOV are assigned to a single host, allowing video, data acquisition, and other specialized cards to be shared across multiple hosts, albeit sequentially.</p> <P> All three vendors have Serial ATA or SAS/SATA drive bays in their I/O expansion chassis. This lets them create a shared direct-attached storage pool allocating logical drives from an SR-IOV RAID controller in the chassis to hosts, which diskless servers can use for boot or other local storage at a lower cost than a boot from the SAN.</p> <P> Virtual I/O may have its biggest impact in the blade server market, where a smaller number of I/O channels helps vendors increase server density. Alliances are starting to form, with IBM integrating NextIO technology in its BladeCenter HT and Dell reselling Xsigo's I/O Director. <P> As a temporary solution until FCoE takes over the world, a lower-cost consolidation point between servers and end-of-row FCoE switches, or a long-term solution, virtual I/O could be worth a look for the more adventurous. On the other hand, those that use new technologies without major-player support always take the risk their virtual I/O system will look like Token Ring in a year or three.</p> <P> <strong><em>Howard Marks is chief scientist at <a href="http://www.naol.com/" target="_blank">Networks Are Our Lives</a>, a consulting firm.</em></strong></p> <P> <span style="font-size:.9em;">Illustration by Jupiterimages</span></p> <P> <center>Continue to the sidebar:<br> <b><a href="http://www.informationweek.com/news/software/server_virtualization/showArticle.jhtml?articleID=217701943">Virtualization Success Is In The Cards</a></b></center><br clear="all" /></p>2009-06-06T00:00:11ZPCI-SIG standards help VMs learn to share.http://www.informationweek.com/news/217701943?cid=RSSfeed_IWK_AuthorsOne key to hosting virtual machines on a single enterprise system is to share PCI cards, like a 10-Gbps Ethernet card, while still writing directly to the card rather than having the hypervisor manage all I/O through the card. After all, sending data through the host operating system, meta device driver, hypervisor, virtual switch, and hypervisor driver -- then having the return data work its way through the same I/O stack -- adds latency. And all those context switches eat CPU cycles.</p> <P> The problem with PCI card sharing is that each virtual machine's operating system believes it has exclusive ownership of all the I/O devices it can reach. A VM will write data to a PCI card and expect data to still be in the card's buffers when it accesses that card again. If another VM accesses that card in the interim, it won't be in the expected state when the first VM accesses it again, and data may be lost.</p> <P> Recognizing these problems, the PCI-SIG industry consortium has developed a pair of I/O virtualization specifications to describe how I/O devices can be shared between host operating systems. The Single Root (SR-IOV) standard specifies how multiple guests on a single server with a single PCI Express (PCIe) controller or root can share I/O devices. A 10-Gbps network interface card that supports SR-IOV can present multiple virtual NICs to VMs with different virtual LAN locations, quality-of-service policies, MAC addresses, etc.</p> <P> Multiple Root IOV (MR-IOV) extends the concept to allow multiple independent systems, with separate PCIe roots, to connect to I/O devices through a switch. MR-IOV systems use the same cables and connectors as other external PCIe devices. Both SR-IOV and MR-IOV require explicit support from I/O cards.</p> <P> Although not every I/O card aimed at servers supports SR-IOV, most 10-Gbps NICs and channel network adapters do, as do some newer Fibre Channel host bus adapters and RAID controllers. MR-IOV hasn't been nearly as widely supported, with just a handful of products available -- and no great demand, either. The trend seems to be for vendors like NextIO and Virtensys to fool SR-IOV cards into doing the things MR-IOV was designed for.</p> <P> <span style="font-size:.9em;">Illustration by Jupiterimages</span></p> <P> <center>Return to the story:<br> <b><a href="http://www.informationweek.com/news/software/server_virtualization/showArticle.jhtml?articleID=217701835">Tech Road Map: Keep An Eye On Virtual I/O</a></b></center></p>2009-04-11T00:00:20ZVendors are lining up to help you trim data bloat. Better choose carefully.http://www.informationweek.com/news/216500085?cid=RSSfeed_IWK_AuthorsIT managers are swamped with data and new mandates to retain it. In fact, data is piling up so fast that just adding more capacity won't solve the problem. Enter storage reduction techniques, ranging from file compression and single-instance storage to data deduplication, which can help a beleaguered IT staff put the proverbial 10 pounds of data in a 5-pound bag.</p> <P> Data deduplication is the latest reduction method to move from secondary storage applications to primary storage systems. There are good reasons for the move, although primary storage isn't always an easy fit for data reduction technology. Large enterprises' primary storage performance requirements are more stringent, especially when it comes to I/O response times and latency. Primary storage systems also have to meet substantially higher availability and reliability standards than backup stores. This makes them leaner, less-target-rich environments for data reduction, but they're also three to 10 times more expensive on a per-gigabyte basis than backup repositories. Small storage reductions can save significantly on space, power, and cooling. There's also a real possibility of performance boosts.</p> <P> <P> <CENTER> <A HREF="http://informationweekreports.com/shared/download.jhtml?id=175300037&cat=iwkr_servers&doc_id=216500085?cid=IWKRPT" target="_blank"><IMG SRC="http://twimgs.com/infoweek/promo/iwkreports.gif" ALT="InformationWeek Reports" WIDTH="278" HEIGHT="25" BORDER="0"></A></CENTER></P> <P> Vendors ranging from enterprise network-attached storage leader NetApp to startups like Ocarina Networks are readying data deduplicating tools to optimize primary storage capacity. With a range of options coming online in the next year or so, from software upgrades to complete NAS systems, now is the time to investigate deduping your primary storage. But keep in mind that your data reduction ratios may be closer to 2-to-1 than 20-to-1. We recommend using conservative data reduction ratios when developing budgets and ROI calculations.</p> <P> <strong>Space And Time</strong><br /> There's more than one way to reduce the amount of space that files occupy on disk. Some data reduction technologies are built into the file system or operating system of a NAS appliance, while others are appliances that can be added to existing filers. These approaches operate either in real time or post-process. Real-time reduction needs the least disk space because it compresses and/or dedupes data as it's written to the share, but it's compute-intensive and can crimp performance. Post-process data reduction happens after data is written to disk. This approach requires enough space to hold both the inflated and deflated versions of the files, but it can be done during off hours, when it's less likely to effect user response time.</p> <P> Deduplication systems find and eliminate duplicate data by dividing files into chunks and looking for chunks that have the same data. The major difference is how they chunk data. The simplest method is to use fixed size chunks like disk blocks.</p> <P> NetApp's Write Anywhere File Layout, or WAFL, builds files as lists of blocks. WAFL calculates a checksum of each block, and stores it with the data, whether by a schedule or by an event like a disk reaching a utilization threshold. Blocks that have the same checksum are compared to see if they contain the same data. If they do, WAFL deletes one block and modifies the metadata of the file where the block resided. WAFL will be released as a free software upgrade later this year.</p> <P> NetApp's approach opts for low overhead instead of high data reduction ratios, so its performance impact should be minimal for the vast majority of applications.</p> <P> Deduplication using variable block size is more complicated but can identify duplicate data in the body of files saved elsewhere. GreenBytes' ZFS+ adds a real-time variable size block to Sun's open source ZFS file system. GreenBytes' Cypress NAS appliance, based on Sun's X4540 storage server, uses variable block sizes to deliver 800-MBps performance -- in part through clever use of flash SSDs to store hash lookup tables and logs. GreenBytes' appliances, priced at $100,000 for 46 TB of raw space, are set for release this summer.</p> <P> <!-- The Essentials --> <div style="margin:0; padding:0 0 10px 10px; width:250px; float:right;"> <div style="border:1px solid #000000; padding:0"> <div style="margin:0; padding:4px; font-size:1.4em; text-align:center; color:#ffffff; background-color:#CC0000;"> <strong>The Essentials</strong> </div> <div style="margin:0; padding:4px; font-size:1.2em; text-align:center; color:#ffffff; background-color:#666666;"> <strong>Primary Storage Reduction</strong> </div> <div style="margin:0; padding:8px; font-size:1.1em;"> <span style="color:#CC0000; font-weight:bold;">1.</span> Consider the options your existing network-attached storage offers. You may get half the bang for none of the bucks. <div style="margin:10px 0 10px 0; padding 0; border-bottom:dashed 1px #000000;"></div> <span style="color:#CC0000; font-weight:bold;">2.</span> Primary storage isn't as readily deduplicated as backups are. Plan for 3-to-1 or 4-to-1, not 20-to-1, reductions. <div style="margin:10px 0 10px 0; padding 0; border-bottom:dashed 1px #000000;"></div> <span style="color:#CC0000; font-weight:bold;">3.</span> Deduplication and large read caches can speed up some apps, including hosting virtual servers. <div style="margin:10px 0 10px 0; padding 0; border-bottom:dashed 1px #000000;"></div> <span style="color:#CC0000; font-weight:bold;">4.</span> Try deduping less-performance-critical data such as user home directories first. <div style="margin:10px 0 10px 0; padding 0; border-bottom:dashed 1px #000000;"></div> <span style="color:#CC0000; font-weight:bold;">5.</span> Deleting or archiving obsolete data will free more space than deduplicating it. </div> </div> </div> <!-- / The Essentials --> Riverbed Technology's Atlas, due next year, puts an Atlas appliance or a redundant pair between the network and any CIFS/NFS file server to deduplicate data in real time. WAN bandwidth is even scarcer than disk space, so Riverbed uses small variable block sizes for high deduplication ratios. Its price has not been announced.</p> <P> Deduplicating frequently accessed data, such as virtual machine images, changes the disk access pattern from reads spread across a volume to accesses of the one deduplicated copy. If the file server has sufficient cache, this replaces many disk I/Os with cache reads. Both NetApp and GreenBytes offer extended read cache options, with GreenBytes offering up to 600GB of flash cache.</p> <P> Where the other dedupe schemes look at files as sets of bits, Ocarina's Online Storage Optimization Solution takes another approach, recognizing common file types and uses different techniques to space-optimize each. Ocarina breaks complex documents like ZIP files or PowerPoint presentations into their component objects. For example, a PowerPoint slide might be deconstructed into a text block, background, logo, photo, and graph, each of which is separately deduped and compressed with algorithms optimized for each data type. An optimizer replaces the files with a series of links to their constituent deduplicated objects. A reader sits between the user and filer, and reassembles data as users access it.</p> <P> Ocarina's road map calls for original equipment manufacturers to integrate its technology into NAS systems. Several large OEMs have committed to the project but none has gone public yet.</p> <P> Howard Marks is chief scientist at Networks Are Our Lives, specializing in data storage, management, and protection. Write to us at <a href="mailto:iweekletters@techweb.com">iweekletters@techweb.com</a>.</p> <P> <center>Continue to the sidebar:<br> <b><a href="http://www.informationweek.com/news/storage/systems/showArticle.jhtml?articleID=216500167">Know Your Options</a></b></center></p>2009-04-11T00:00:10ZOther ways to store more data on primary systems include compression and single-instance storage.http://www.informationweek.com/news/216500167?cid=RSSfeed_IWK_AuthorsDeduplication isn't the only data reduction plan in town. Compression and single-instance storage provide other tried-and-true ways to store more data on primary systems.</p> <P> Lossless compression using the LZW algorithm, or one of its many descendants, dates back to the early 1990s when Stacker appeared to compress DOS, and therefore Windows 3.x, disks. Even in those dark ages of 200-MHz Pentium Pros and 40-MBps SCSI drives, compression almost always improved I/O performance by forcing more data through the disk channel bottleneck.</p> <P> Today's processors are 100 times faster than those of the early '90s, but the disk I/O channel is only 16 times bigger than it was then. So we think fears of compression negatively impacting performance are overblown at best.</p> <P> NTFS compression has been a standard feature of Windows servers since the beginning, enabling administrators to compress data on a folder-by-folder basis. Conversely, enterprise NAS vendors have until recently avoided compression because of fears it might slow performance.</p> <P> Storwize's STN-6000 appliances sit between the user network and CIFS or NFS servers. They intercept requests and compress or decompress the data payload using a version of LZ compression tuned for random I/O, then pass it on to the filer. Storwize claims typical users get 75% (or 3-to-1) data compression, with especially compressible data getting up to 10-to-1, and most applications also see improved performance.</p> <P> <strong>Single-Instance Storage</strong><br /> Single-instance storage identifies multiple copies of the same file in a file system and replaces all but one with references to a single copy of the data. The file system manages updates by keeping track of how many files refer to a single set of data, knowing when users create new, modified copies, and deleting data after the last file referencing it is deleted. Single instance works well on user home directories and similar sets of files because many users will save the same e-mail attachment or scanned delivery menu in their home folders.</p> <P> EMC's Celerra NAS, which ranges in price from $20,000 to more than $100,000, uses both single-instance storage and data compression in its data reduction toolkit, and addresses possible performance impacts by only using single instancing and/or compressing inactive files. Admins set policies to apply single-instance storage and compression based on the file's last accessed time stamp and schedule a task to apply the policies.</p> <P> Windows Storage server uses NTFS's real-time compression and, like Celerra, provides single-instance storage as a scheduled post-process.</p> <P> <center>Return to the story:<br> <b><a href="http://www.informationweek.com/news/storage/showArticle.jhtml?articleID=216500085">Deduplication Joins The Primary Storage Reduction Fray</a></b></center></p>2009-02-28T00:00:04ZKnow your systems to find the failsafe failover that works best.http://www.informationweek.com/news/214600295?cid=RSSfeed_IWK_AuthorsEnterprises of all sizes demand 24/7 application delivery. Server failures, maintenance downtime, and acts of nature are no excuse. If keeping key applications online is your job, you should consider yourself lucky: You have more options for keeping apps up than ever before.</p> <P> With the tools available today, organizations have few excuses -- not even budgetary ones -- for relying on the hours-long process of manually restoring mission-critical apps from backup.</p> <P> <P> <CENTER> <A HREF="http://informationweekreports.com/shared/download.jhtml?id=174800002&cat=iwkr_apo&doc_id=214600295?cid=IWKRPT" target="_blank"><IMG SRC="http://twimgs.com/infoweek/promo/iwkreports.gif" ALT="InformationWeek Reports" WIDTH="278" HEIGHT="25" BORDER="0"></A></CENTER></P> <P> Application failover approaches run the gamut, from basic clustered server "ping and a prayer" software to complete virtualized systems and application-specific schemes. Finding the one that's right for you will involve more than a glance at the price tag, which runs from $1,500 to $10,000-plus per protected server. You'll also need to consider ease of use, speed of failover, bandwidth consumption, and how much data is at risk.</p> <P> When most system administrators look to improve application availability, they start with server clusters. Failover clustering has been available in Windows Server's Enterprise Editions since Windows NT 4 was state of the art in the mid 1990s, but it developed a well-deserved reputation for being finicky.</p> <P> Windows clusters used shared storage, which of course made the storage subsystem a single point of failure, until Windows Server 2008 was released. Microsoft insisted on only integrated server and storage solutions, so users that had Hewlett-Packard servers and EqualLogic storage, for example, were out of luck when it came to support. Most significantly, applications had to be cluster-aware to smoothly fail over from one server node to another.</p> <P> Even before Microsoft added clustering to Windows itself, vendors like Double-Take Software released solutions that combined data replication, which eliminates storage as a single point of failure, with automatic failover. Early versions of these products required a lot of setup and tweaking, including installing the OS and applications on both servers. However, the current crop, such as SteelEye's LifeKeeper, CA/XOsoft's WANsync, NeverFail's Continuous Protection Suite, and, of course, Double-Take, can clone a production server to the standby server, both speeding setup and ensuring the servers are similarly configured. And some of these offerings support Linux clustering as well as traditional Windows clusters.</p> <P> In a generic cluster or high-availability system, the failover server, or servers, monitor the primary host by exchanging heartbeat messages across the network (see diagram, "<a href="#diagram">Two Ways To Keep Apps At Your Service</a>"). If the primary host doesn't respond within a given period of time, the standby server assumes the primary host's identity and starts processing data in its place.</p> <P> This method can prevent data loss due to a complete failure of the primary host and allows manual failovers for patching and other server maintenance, but it can't detect more subtle failures of services and daemon processes. Vendors including SonaSoft and Marathon sell more app-aware offerings, which check the state of services or connect directly to applications to ensure they're running.</p> <P> Products also use different methods to allow a standby server to assume the identity of a production server in the event of a failure. The simplest way is to assume the production server's IP address and start appropriate services. A more sophisticated approach used by NeverFail and others is to hide the standby server behind an internal firewall to prevent users from accessing it until it's called on to take on the primary server role. At the top end of the product spectrum, Marathon's EverRun runs the primary and standby servers in lockstep in a virtual environment. Each server processes all data, but users access only the primary one. The backup server waits in the wings until something goes wrong.</p> <P> <a name="diagram"></a> <center><div align="center" style="width:440px; background-color:#7696ca; text-align:center; margin:0; padding:0;"><IMG SRC="http://twimgs.com/infoweek/1222/222TS_failover_diagram.gif" ALT="diagram: Two Ways To Keep Apps At Your Service" WIDTH="440" HEIGHT="162" HSPACE="0" VSPACE="0" BORDER="0" /> <div style="margin:0; padding:8px 10px; text-align:center; font-size:1em; color:#FFFFFF; font-weight:bold; line-height:1em;">In a high-availability cluster (left), a data center's standby server can adopt the primary server's IP address and identity when it doesn't respond over the heartbeat link. Effective disaster recovery (right) demands a more sophisticated scheme: In the event of server failure, the data must fail over to the standby server, often in a remote facility, via wireless LAN.</div> </div></center></p><strong>Virtually Constant</strong><br /> Just as in data center management, server virtualization has enabled administrators to greatly improve availability.' The most obvious impact is that virtualizing standby servers breaks the expensive one-to-one relationship between production and standby servers, thus reducing the cost of providing standby servers.' Because all virtual servers on the same virtualization platform look identical to the guest OS, driver and other related hardware issue are eliminated.</p> <P> Organizations can also quickly provision servers in data center high-availability systems at the virtual server host. VMware High Availability, Microsoft Clustering of Hyper-V, and Linux failover solutions for Xen all protect guest servers from host failures and allow host maintenance without significant guest downtime. Marathon's EverRun for Hyper-V and Xen can extend host protection to true multisite disaster recovery as well.</p> <P> While VMware's Site Recovery Manager (SRM) does require some scripting, it also provides for site-to-site failover of virtual servers across a variety of applications and guest operating systems. SRM relies on storage arrays to replicate the data from site to site and array manufacturers have to write an adapter to enable SRM to manage the replication process.</p> <P> Another recent trend in application availability has been the development of high-availability and disaster-recovery solutions that are not only application-aware but also operate at the application layer.' A general-purpose solution replicates file- or block-level writes to the primary host's storage to a standby storage system.</p> <P> Regardless of whether the replication is done using software in the primary and standby hosts or by the storage system itself, the application's database is being managed by the primary host with the standby's copy of the application sitting idly by. When the failover occurs, the application starts on the standby server and mounts its "crash-consistent" copy of the database. (Crash consistent is the industry euphemism for a database that's as consistent as it would be when the server crashes -- or, in plain English, not consistent at all since some number of transactions were assumed to be in the middle of being processed when the server crashed.) Therefore, the first thing the server has to do is a quick consistency check to roll back the transactions that were in progress when the crash occurred. This process usually takes just a minute or two but can occasionally leave the server unavailable to users for several hours as the database is checked and reindexed, especially if the crash occurs in the middle of a database defragmentation.</p> <P> Application-specific solutions replicate transactional data to a standby server where the running application applies the transaction to its copy of the database. This approach has several advantages. First, because the backup server is running the application, it usually doesn't take long to fail over to the backup, start the application processes, and mount the database. Second, posting completed transactions prevents many sources of database corruption, such as those caused by malware on the primary host, or storage system I/O errors, from propagating to the backup server.</p> <P> The secondary server can also be used as a data source for operations like backup, archiving, and reporting, allowing these processes to run anytime without affecting users.</p> <P> Replicating transactional data also reduces the amount of data that must be sent between primary and secondary data stores. Modern databases write data to transaction log files and then, when the transaction is complete, to the on-disk database. Solutions that replicate storage data must replicate the writes to both the transaction log and database, whereas transaction-based solutions only have to send any given transaction across the line once.</p> <P> <!-- The Essentials --> <div style="margin:0; padding:0 0 10px 10px; width:250px; float:right;"> <div style="border:1px solid #000000; padding:0"> <div style="margin:0; padding:4px; font-size:1.4em; text-align:center; color:#ffffff; background-color:#CC0000;"> <strong>The Essentials</strong> </div> <div style="margin:0; padding:4px; font-size:1.2em; text-align:center; color:#ffffff; background-color:#666666;"> <strong>Finding Failover That Fits</strong> </div> <div style="margin:0; padding:8px;"> <span style="font-size:1.2em; color:#CC0000; font-weight:bold;">Server clusters:</span><br /> Work best in smaller environments with basic failover needs Application-specific software: Tailors failover chores to individual apps; good choice when only a few apps are needed 24/7 <div style="margin:10px 0 10px 0; padding 0; border-bottom:dashed 1px #000000;"></div> <span style="font-size:1.2em; color:#CC0000; font-weight:bold;">Application-aware software:</span><br /> Best for organizations that need fast remediation of failures that are more subtle than complete server crashes <div style="margin:10px 0 10px 0; padding 0; border-bottom:dashed 1px #000000;"></div> <span style="font-size:1.2em; color:#CC0000; font-weight:bold;">Virtual failover systems:</span><br /> Make the most sense for larger enterprises with multiple critical apps on many servers </div> </div> </div> <!-- / The Essentials --> Because each application server performs the task on its own database, this approach also avoids the time-consuming chore of replicating disk writes created when applications perform internal database defrag -- which Exchange 2003 performs nightly -- and other housekeeping.</p> <P> Many application-specific failover vendors have focused on Exchange, in no small part because it has so many moving parts and interconnections to Active Directory and other network services. Software like Cemaphore's MailShadow OnSite and SonaSoft's SonaSafe for Exchange capture data from the Exchange server using the native Exchange MAPI protocol and transfer it to a running Exchange server; one backup server can provide protection for several source servers and with SonaSafe production servers in different offices can back each other up. To fail over, they run a script that updates the user's mailbox location data in active directory; users then connect to their mailboxes on the standby server.</p> <P> Teneros' Application Continuity Appliance packages the MAPI data acquisition, failover, and standby Exchange server in an appliance that's positioned inline between the users and the production Exchange server. It will also asynchronously replicate data to an additional appliance at a remote site for disaster-recovery purposes.'</p> <P> Microsoft has even dipped its toe in the water with disaster recovery features for Exchange 2007: Cluster continuous replication (CCR) for high availability and standby continuous replication (SCR), which ship transaction log files from primary to secondary servers to keep the database up to date. SCR requires significant manual intervention, or scripting, to bring up the standby server, but it shows promise. CCR relies on Windows clustering for failover and is limited to having all the systems on the same subnet.</p> <P> The downside to application-level failover is that, in the event of a primary server failure, at least a few transactions will be lost in transition, posted to the primary server but not replicated to the standby. So while these solutions improve recovery time, they can also negatively affect recovery points.</p>2009-02-21T00:00:00ZVirtual machine backup software is evolving rapidly, and there's no one best tool.http://www.informationweek.com/news/214501909?cid=RSSfeed_IWK_AuthorsMany organizations have virtualized their test, development, and low-duty-cycle application servers, freeing up rack space and reducing the power and cooling load in the data center. Now, as these organizations get ready to virtualize the servers that hold their most vital, high-traffic data, they're wondering: "How do we back these things up?" </P> <P> There are a couple of paths that lead to easy, reliable virtual server backups, from the obvious to the innovative. Administrators can conduct image backups that make total system backups and restores fast and relatively painless--but they must be repeated often to keep up with virtual system changes. The alternative, agent-based backups on each host, provides file cataloging and indexing, direct backup and restore from tape, and individual item restore from databases like Exchange--but requires significant system resources and careful management to avoid agent sprawl. </P> <P> <P> <CENTER> <A HREF="http://informationweekreports.com/shared/download.jhtml?id=174600005&cat=iwkr_virtualization&doc_id=214501909?cid=IWKRPT" target="_blank"><IMG SRC="http://twimgs.com/infoweek/promo/iwkreports.gif" ALT="InformationWeek Reports" WIDTH="278" HEIGHT="25" BORDER="0"></A></CENTER></P> <P> While weighing the pros and cons of each approach, administrators also have to choose their server virtualization backup tools. Like server virtualization platforms themselves, virtual machine backup software is evolving rapidly. Vendors are working to provide the same level of consistency and granularity regardless of whether VMs are backed up via host agents, guest agents, or a centralized proxy like VMware's Consolidated Backup. Linux and Solaris file-level restores are also on the virtual menu. </P> <P> Still, there's no one best backup tool for virtual servers. New software designed specifically for virtual systems might be a good choice for backing up servers that need full data restoration more than partial restores; tried-and-true workhorse tools may be better when data needs to be stored for more than a few days. </P> <P> <!-- The shortlist --> <div style="margin:0; padding:0 0 10px 10px; width:280px; float:right;"> <div style="border:1px solid #ffCC33;; padding:10px;"> <div style="margin:0 0 8px 0; padding:0; text-align:center;"> <strong><span style="font-size:1.5em;">The</span> <span style="color:#ffCC33; font-size:1.5em;">Short List</span></strong> </div> <div style="margin:0 0 8px 0; padding:3px; text-align:center; background-color:#ffCC33;; color:#ffffff; font-size:1.1em;"><strong>VIRTUAL SERVER BACKUP VENDORS</strong> </div> <TABLE border="0" width="255" bgcolor="#000000" cellpadding="2" cellspacing="1"> <P> <TR bgcolor="#ffffff"> <TD >&nbsp;</TD> <TD><span style="color:#ffCC33; font-size:.9em;"><strong>Agent On Guest OS</strong></span></TD> <TD><span style="color:#ffCC33; font-size:.9em;"><strong>Agent/ Backup Software On Host</strong></span></TD> <TD><span style="color:#ffCC33; font-size:.9em;"><strong>Consli- dated Backup</strong></span></TD> </TR> <P> <TR bgcolor="#ffffff" align="center"> <TD valign="top"><span style="font-family:verdana; font-size:11px;"><strong>Atempo</strong></TD> <TD valign="top"><span style="font-family:verdana; color: #CC0000; font-size:17px;"><strong>X</strong></span></TD> <TD valign="top"><span style="font-family:verdana; color: #CC0000; font-size:17px;"><strong>X</strong></span></TD> <TD valign="top"><span style="font-family:verdana; color: #CC0000; font-size:17px;"><strong>X</strong></span></TD> </TR> <TR bgcolor="#ffffff" align="center"> <TD valign="top"><span style="font-family:verdana; font-size:11px;"><strong>BakBone</strong></TD> <TD valign="top"><span style="font-family:verdana; color: #CC0000; font-size:17px;"><strong>X</strong></span></TD> <TD valign="top"><span style="font-family:verdana; color: #CC0000; font-size:17px;"><strong>X</strong></span></TD> <TD valign="top"><span style="font-family:verdana; color: #CC0000; font-size:17px;"><strong>X</strong></span></TD> </TR> <TR bgcolor="#ffffff" align="center"> <TD valign="top"><span style="font-family:verdana; font-size:11px;"><strong>CA</strong></TD> <TD valign="top"><span style="font-family:verdana; color: #CC0000; font-size:17px;"><strong>X</strong></span></TD> <TD valign="top"><span style="font-family:verdana; color: #CC0000; font-size:17px;"><strong>X</strong></span></TD> <TD valign="top"><span style="font-family:verdana; color: #CC0000; font-size:17px;"><strong>X</strong></span></TD> </TR> <TR bgcolor="#ffffff" align="center"> <TD valign="top"><span style="font-family:verdana; font-size:11px;"><strong>CommVault</strong></TD> <TD valign="top"><span style="font-family:verdana; color: #CC0000; font-size:17px;"><strong>X</strong></span></TD> <TD valign="top"><span style="font-family:verdana; color: #CC0000; font-size:17px;"><strong>X</strong></span></TD> <TD valign="top"><span style="font-family:verdana; color: #CC0000; font-size:17px;"><strong>X</strong></span></TD> </TR> <TR bgcolor="#ffffff" align="center"> <TD valign="top"><span style="font-family:verdana; font-size:11px;"><strong>EMC</strong></TD> <TD valign="top"><span style="font-family:verdana; color: #CC0000; font-size:17px;"><strong>X</strong></span></TD> <TD valign="top"><span style="font-family:verdana; color: #CC0000; font-size:17px;"><strong>X</strong></span></TD> <TD valign="top"><span style="font-family:verdana; color: #CC0000; font-size:17px;"><strong>X</strong></span></TD> </TR> <TR bgcolor="#ffffff" align="center"> <TD valign="top"><span style="font-family:verdana; font-size:11px;"><strong>IBM</strong></TD> <TD valign="top"><span style="font-family:verdana; color: #CC0000; font-size:17px;"><strong>X</strong></span></TD> <TD valign="top"><span style="font-family:verdana; color: #CC0000; font-size:17px;"><strong>X</strong></span></TD> <TD valign="top"><span style="font-family:verdana; color: #CC0000; font-size:17px;"><strong>X</strong></span></TD> </TR> <TR bgcolor="#ffffff" align="center"> <TD valign="top"><span style="font-family:verdana; font-size:11px;"><strong>PHD Technologies</strong></TD> <TD valign="top"><span style="font-family:verdana; font-size:11px;">&nbsp;</span></TD> <TD valign="top"><span style="font-family:verdana; font-size:11px;">&nbsp;</span></TD> <TD valign="top"><span style="font-family:verdana; color: #CC0000; font-size:17px;"><strong>X</strong></span></TD> </TR> <TR bgcolor="#ffffff" align="center"> <TD valign="top"><span style="font-family:verdana; font-size:11px;"><strong>Symantec</strong></TD> <TD valign="top"><span style="font-family:verdana; color: #CC0000; font-size:17px;"><strong>X</strong></span></TD> <TD valign="top"><span style="font-family:verdana; color: #CC0000; font-size:17px;"><strong>X</strong></span></TD> <TD valign="top"><span style="font-family:verdana; color: #CC0000; font-size:17px;"><strong>X</strong></span></TD> </TR> <TR bgcolor="#ffffff" align="center"> <TD valign="top"><span style="font-family:verdana; font-size:11px;"><strong>Veeam</strong></TD> <TD valign="top"><span style="font-family:verdana; font-size:11px;">&nbsp;</span></TD> <TD valign="top"><span style="font-family:verdana; font-size:11px;">&nbsp;</span></TD> <TD valign="top"><span style="font-family:verdana; color: #CC0000; font-size:17px;"><strong>X</strong></span></TD> </TR> <TR bgcolor="#ffffff" align="center"> <TD valign="top"><span style="font-family:verdana; font-size:11px;"><strong>Vizioncore</strong></TD> <TD valign="top"><span style="font-family:verdana; font-size:11px;">&nbsp;</span></TD> <TD valign="top"><span style="font-family:verdana; font-size:11px;">&nbsp;</span></TD> <TD valign="top"><span style="font-family:verdana; color: #CC0000; font-size:17px;"><strong>X</strong></span></TD> </TR> </TABLE> <P> </div></div> <P> The most obvious way to back up a virtualized server is to install an agent for your existing system on each virtual server and use the same procedures you use now. Backing up virtual servers with locally installed agents has its advantages, not the least of which is familiarity. Using local agents generally provides the most granular backups and is the only way to restore individual items from databases, such as message-level restores. </P> <P> However, using an agent on each virtual server can create a substantial load on the host CPU and network connections, not to mention the cost and effort associated with installing and maintaining all those agents. Also, because the agent runs in the virtual machine, this technique can't back up VMs that are shut down. </P> <P> <strong>Host-Level Backup</strong></P> <P> The other option is to back up virtual server hosts, and therefore guest VMs, through an agent installed in the host's operating system or the virtualization platform's service console. It makes sense: If you have 16 virtual servers running on each host, backing up the hosts rather than the individual guests, that will be one-sixteenth as many backup jobs to manage and one-sixteenth the number of agents to install and update. </P> <P> Most basic backup applications take a snapshot of the logical drives that hold a VM's virtual disk and configuration files. Those snaps are then backed up as files in the host. </P> <P> A host operating system backs up the VMs and their virtual drives as a set of files. However, VM volumes with direct mappings to SAN logical disks via iSCSI or Fibre Channel can't be backed up this way because they're not visible to the host operating system. </P> <P> Most organizations start by consolidating physical servers to virtual ones, then use their hypervisor's live-motion features to move servers as user demand varies. However, with host-level backup, admins must modify backup schedules to account for every move a VM makes, or risk not backing up the migrated VM. In these cases, companies may need to install additional job manager/scheduler software that can keep track of hosts and the VMs on those hosts. Most virtual server backup vendors don't provide this capability. </P> <P> <strong>The Scramble Is On</strong></P> <P> During the past couple of years, a wide range of software vendors has created backup apps that capitalize on VMware's built-in snapshot capability, either directly or via proxies to run scheduled disk-to-disk backups of VMs and their data. </P> <P> For example, PHD Technologies esXpress and products like it create hot image backups of virtual machines through virtual backup appliance VMs on each VMware ESX host. Once the snapshot file is created, esXpress can compress and optionally encrypt the files, and save them to a VM file system, FTP, SSH, or CIFS server across the network. </P> <P> <!-- The Essentials --> <div style="margin:0; padding:0 0 10px 10px; width:290px; float:right;"> <div style="border:1px solid #3399ff; padding:12px;"> <div style="margin:0 0 8px 0; padding:0; text-align:center;"> <strong><span style="font-size:1.5em;">The</span> <span style="color:#3399CC; font-size:1.5em;">Essentials</span></strong> </div> <div style="margin:0 0 8px 1; padding:3px; text-align:center; background-color:#3399ff; color:#ffffff; font-size:1.4em;"><strong>VIRTUAL BACKUP METHODS</strong> </div> <TABLE border="0" width="210" bgcolor="#000000" cellpadding="2" cellspacing="1"> <P> <TR bgcolor="#ffffff"> <TD >&nbsp;</TD> <TD><span style="color:#3399CC; font-size:1em;"><strong>Pros</strong></span></TD> <TD><span style="color:#3399CC; font-size:1em;"><strong>Cons</strong></span></TD> </TR> <P> <TR bgcolor="#ffffff"> <TD valign="top"><span style="font-family:verdana; font-size:11px;"><strong>Agent On Guest OS</strong></TD> <TD valign="top"><span style="font-family:verdana; font-size:11px;">Same backup process as physical servers; high backup/restore granularity</span></TD> <TD valign="top"><span style="font-family:verdana; font-size:11px;">High load on host server; many agents to maintain</span></TD> </TR> <TR bgcolor="#ffffff"> <TD valign="top"><span style="font-family:verdana; font-size:11px;"><strong>Agent/Backup Software On Host</strong></TD> <TD valign="top"><span style="font-family:verdana; font-size:11px;">Fewer agents to maintain</span></TD> <TD valign="top"><span style="font-family:verdana; font-size:11px;">High load on reduced restore granularity</span></TD> </TR> <TR bgcolor="#ffffff"> <TD valign="top"><span style="font-family:verdana; font-size:11px;"><strong>Conslidated Backupt</strong></TD> <TD valign="top"><span style="font-family:verdana; font-size:11px;">Fast LAN-free backups, integration with existing backup apps</span></TD> <TD valign="top"><span style="font-family:verdana; font-size:11px;">Reduced restore granularity; requires scripting or software</span></TD> </TR> </TABLE> <P> </div></div> <!-- / The Essentials --> PHD esXpress ranges from free to $1,995 per ESX host. The free edition is limited to two simultaneous backups at 13 Mbps each; other editions add features, performance, and more simultaneous backups.</P> <P> Veeam Backup and Vizioncore vRanger Pro both perform disk-to-disk backups of VMware ESX servers, and both can ensure application-consistent backups for Windows VMs. VRanger is priced at $499 per ESX host processor with volume discounts.</P> <P> Like comparable products, Veeam Backup performs a full backup the first time it encounters a VM and incremental backups thereafter. It's priced at $495 per ESX host processor. </P>2009-02-06T11:56:40ZIn yet another chapter in our continuing series bringing further embarrassment to poor souls that were foolish enough to not have a viable backup plan, we have the sad tale of blog hosting firm JournalSpace. It managed to survive six years using RAID as a substitute for backups. But then data corruption struck and business failure soon followed.http://www.iweek-interim.com/news/229207608?cid=RSSfeed_IWK_AuthorsIn yet another chapter in our continuing series bringing further embarrassment to poor souls that were foolish enough to not have a viable backup plan, we have the sad tale of blog hosting firm JournalSpace. It managed to survive six years using RAID as a substitute for backups. But then data corruption struck and business failure soon followed.Best known for hosting the blog of a former Delta Airlines flight attendant who was fired for posting photos of herself showing a little leg (and not much more) inside the cabin of a plane, JournalSpace's IT guy had scripts to protect the PHP scripts on the front-end servers but relied on mirrored drives to protect the database server. <P> JournalSpace employees described the admin in question as having a hobby of telling other people how smart he was. We've all run into this type in the IT biz. Those who can do, those who can't may teach (although so may those who can), and only those that can't do or teach brag. <P> After accusations of theft from the company and sabotage: Poof! No database. No database, no blogs. <P> Some users have resorted to Google's online cache to recover their data. If you had a JournalSpace blog and want to try recovering data via Google, see <a href="http://andrewtheart.blogspot.com/2009/01/how-to-recover-mostall-of-your.html">this blog post</a>. <P> After the disaster, the owners of JournalSpace.com sold the domain to new owners who resurrected the site, but not the data. <P> Lessons for today: <P> 1. RAID isn't a substitute for backups. 1a. Neither are snapshots. 2. Bloggers should back up their content themselves, as hosting providers can only be trusted so far. 3. Never underestimate the danger of a disgruntled employee.2009-01-27T11:07:53ZWhite House Press Secretary Robert Gibbs announced at a 1:45 p.m. press briefing yesterday that he was unable to send out the customary week-ahead memo as the White House e-mail system was "not working so well." D.C. reporters got their next e-mail from the White House around 8:30 this morning indicating that the outage lasted most of a day.http://www.iweek-interim.com/news/229208072?cid=RSSfeed_IWK_AuthorsWhite House Press Secretary Robert Gibbs announced at a 1:45 p.m. press briefing yesterday that he was unable to send out the customary week-ahead memo as the White House e-mail system was "not working so well." D.C. reporters got their next e-mail from the White House around 8:30 this morning indicating that the outage lasted most of a day.Combine this with the last administration's neglecting, or refusing, depending on your political point of view, to install an e-mail archiving solution and "oops" losing backups of many messages that may shed light on historical events and you have to start worrying about the competence of the IT people over there. <P> Given our first Hawaiian president's well-known predilection to use his BlackBerry and the campaign's fine use of e-mail and I'm hoping this glitch is just the result of systems left over from the Bush administration. Even so, I'm going to give some unsolicited advice: <P> <b>1)</b> Get an Exchange archiving solution. The Presidential Records Act requires retention and a good archiving solution is the right way to get it. Get Symantec Enterprise Vault, Mimosa NearPoint, or any of dozens of others. <P> <b>2)</b> Get a high-availability solution such as Cemaphore's MailShadow or products from NeverFail, Marathon, etc., etc., etc. <P> My friend Cemaphore CEO Tyrone Pike even mentioned that they'd be glad to donate MailShadow to the White House as a patriotic act, and, I suspect, to get a little publicity. I'm sure I can convince an archiving vendor to chip in, too. <P> White House geeks can e-mail me at <a href="mailto:hmarks@nwc.com">hmarks@nwc.com</a> for any assistance they may need.2008-12-20T00:01:30ZIn this InformationWeek Analytics report, we profile new technologies that can protect the business without breaking the bank.http://www.informationweek.com/news/212501197?cid=RSSfeed_IWK_AuthorsDisasters happen, and when they do, IT had better be prepared, since businesses depend on information and the technology that manages it. For midsize companies, however, planning and equipping for disasters has been problematic. Where large enterprises have an array of specialized disaster-recovery systems from which to choose, and small businesses often can make do with ad hoc measures, midsize companies frequently have been caught in the middle--not able to afford big-bucks systems, yet needing more than just sending tapes off-site.</p> <P> Whether it's the lingering memory of disasters like Hurricane Katrina, or a recent brownout that sent servers belly-up, or regulatory requirements requiring signatures from the top brass, CEOs are increasingly aware of the problem. At midsize companies, 79% of respondents to our <i>InformationWeek</i> Analytics survey say they have a business-continuity plan in place, and 42% cite lack of C-level awareness as a barrier to disaster-recovery planning. However, companies vary widely in how quickly they can recover: 35% say they'd be back in a few days; 28% say it would be a few weeks.</p> <P> Increasing midsize business preparedness means using new technologies that are changing the game for disaster recovery, but it also requires spending time classifying applications and managing expectations for the recovery process. The fundamental issues are which apps are required to run the business, how much data can be lost, and how long you can wait to restore functionality. (For a more extensive discussion, <a href="http://data-protection.informationweek.com">see our full report</a>, it includes discussions of distance issues, critical personnel matters, and more.)</p> <P> For example, a CAD system might be very important to a business, but perhaps nightly tape backups and a week to rebuild servers is sufficient in the face of a flood or tornado. Engineers can re-create any work that's lost, and after a local disaster, they'll be attending to their families' needs before digging back into the company's latest designs. Geography also can make a big difference. A company with engineers in more than one location will prefer to replicate data between the groups, and while a disaster will certainly reduce overall productivity, little information will be lost.</p> <P> <strong>CRITICAL METRICS TO MANAGE</strong><br /> Two key statistics to plan around are the recovery point objective and the recovery time objective, or RPO and RTO. The RPO speaks to how much data is acceptable to be irretrievably lost. In the CAD example above, as much as a day's worth of data may be lost, since backups are only done nightly. For an e-mail system, a company might be willing to lose only an hour's worth of data, and for a transaction-processing system, it might not be acceptable to lose any data.</p> <P> Recovery time objective speaks to the time required to bring a system back up. While it may be OK to let the CAD system wait a week, the e-mail system is critical. Here, the system itself actually has a different RTO than does its data. Employees can get lots of useful work done if e-mail's running--two-thirds of companies have messaging in their disaster plans--even without back e-mail files. Transaction-processing systems may need to be back in operation in minutes.</p> <P> IT teams can't determine RPO and RTO of the systems in a vacuum; they need to survey users and consult line-of-business managers and executives on the value of lost time and data. The different RPOs and RTOs must be considered when designing an application, not just in how it's backed up. For instance, a transaction-processing e-commerce system with very low RPOs and RTOs might be best run in a co-location facility that has the redundant Internet and power grid connections, generators, and adequate security, items many midsize businesses can't afford in their own data centers. Some midsize companies, particularly in places such as the San Francisco area and the Gulf Coast, use co-location facilities for part of their data center needs and use their local sites as the disaster-recovery backup.</p> <P> <div style="margin:0; padding:0 0 10px 10px; width:320px; float:right;"><img src="http://twimgs.com/infoweek/1216/216ID3_DisasterRecovery_chart21.gif" alt="chart: What are the top barriers to adoption of a business continuity plan?" width="310" height="335" hspace="0" vspace="0" border="0"> </div> If companies want to reduce RPOs from hours to minutes, they need to replicate write requests in real time to a duplicate store at their disaster-recovery sites. There are various replication products available, each designed to provide the best balance of cost, bandwidth requirements, RPO, and ease of management for a given set of uses. There are three commonly used data replication methods: synchronous, asynchronous, and snapshot.</p> <P> Synchronous replication systems duplicate each write request, sending to both the primary and secondary data store. They wait for both stores to complete the write before considering the transaction fully processed. This brings RPO to zero, but the cost is high, and--if not designed properly--such products can substantially slow application performance. These systems typically are used by large organizations and employ dark fiber or some other high-performance network.</p> <P> Asynchronous replication also drives RPO toward zero, but does so without waiting for every write to be acknowledged from both local and remote data stores. It often includes management and optimization techniques to minimize bandwidth use.</p> <P> Snapshot technologies work at the storage level, transferring modified blocks of data at set intervals. These systems are more bandwidth-friendly and often work without application awareness. While SAN-based systems are the gold standard for data replication, they often cost more than midsize organizations can afford.</p> <P> Increasingly, host-based and appliance-based replication, or systems based on server virtualization, are attractive to midsize companies. Not only is the price lower, but virtualization can greatly reduce the time and equipment needed to bring up a disaster recovery site.</p> <P> <center><img src="http://twimgs.com/infoweek/1216/216ID3_DisasterRecovery_chart28.gif" alt="chart: What parts of your IT infrastructure are covered by your business continuity or disaster recovery plan? " width="314" height="428" hspace="0" vspace="0" border="0"></center></p> <P> <div style="margin:0; padding:6px;; font-size:1.1em; font-weight:bold; color:#ffffff; text-align:center; background-color:#4b6a84;"><strong>&gt;&gt; See the full report at <a href="http://data-protection.informationweek.com" style="color:#f1de7e;">data-protection.informationweek.com</a> &lt;&lt;</strong></div></p>2008-12-11T12:08:34ZWhile online backup vendors like SpiderOak are offering discounts for displaced Xdrive users and AOL lists Dropbox, Carbonite and Box.net along with Elephant Drive as replacements for Xdrive users I agree with Matt K Olsen who commented on my <a href="http://www.informationweek.com/blog/main/archives/2008/11/aol_throws_xdri.html">last blog post</a> on this issue the Elephant Drive was the best replacement for Xdrive users.http://www.iweek-interim.com/news/229208633?cid=RSSfeed_IWK_AuthorsWhile online backup vendors like SpiderOak are offering discounts for displaced Xdrive users and AOL lists Dropbox, Carbonite and Box.net along with Elephant Drive as replacements for Xdrive users I agree with Matt K Olsen who commented on my <a href="http://www.informationweek.com/blog/main/archives/2008/11/aol_throws_xdri.html">last blog post</a> on this issue the Elephant Drive was the best replacement for Xdrive users.Elephant Drive has a backup application like Carbonite or Mozy and secure file sharing like Box.net or Dropbox backed up by a solid infrastructure that stores user data in multiple locations including Elephant's own servers at collocation centers and Amazon's S3. The specialists may have richer user interfaces and more features but Elephant covers the bases. <P> But Xdrive users should flock to Elephant's Trunk Drive feature that like Xdrive lets Windows users map a drive letter to Elephant's cloud storage. <P> Even better Elephant is the only vendor I'm aware of that has a direct data migration offer for Xdrive users. As Elephant's CEO Michael Fisher explained to me "We have an access API and so did XDrive so we just wrote an app that uses both APIs to transfer data from their cloud storage to ours without having to run it through the customer's PC" which could be a big deal for users with a lot of data and/or asymmetrical links that make uploading slow. <P> All an Xdrive user has to do is load the <a href="https://www.elephantdrive.com/online_backup/xdrive_elephantdrive_migration.aspx">migration page</a> and enter their Xdrive userID and password. Elephant even uses Amazon's EC3 infrastructure to run the transfer so there's no shortage of compute power to run the transfer.2008-11-30T21:09:04ZWelcome to the silly season, when marketers decide their company's products make great holiday gifts. As <i>InformationWeek</i>'s Master of Disaster, I get e-mail from all sorts of folks who think I should say nice things about their products in this here blog. Sometimes they try just a little too hard to make their products topical.http://www.iweek-interim.com/news/229208905?cid=RSSfeed_IWK_AuthorsWelcome to the silly season, when marketers decide their company's products make great holiday gifts. As <i>InformationWeek</i>'s Master of Disaster, I get e-mail from all sorts of folks who think I should say nice things about their products in this here blog. Sometimes they try just a little too hard to make their products topical.Today I opened an e-mail from emergancylifeline.com claiming to offer "Great Gift Ideas." Turns out it's selling survival kits. <P> Much as keeping some food, water, and a first aid kit around the house and/or data center might be a good idea in earthquake or hurricane country, a great gift idea it's not. <P> While I smiled and wrote thank you notes to Aunt Charlotte for the itchy wool socks she knitted for me throughout my childhood, I really wanted an Erector Set. <P> I just can't see saying "Oh, boy, survival food bars! Just want I want for Hanukkah."2008-11-28T05:45:16ZAt least the T-shirts they're giving away at <a href="http://backthefup.net/">backthefup.net</a> do. As the site says, "Screw Klondike&#174; Bars, What Would You Do For A Back The F:\ Up T-Shirt?" Turns out what you have to do is something that promotes EMC's Mozy backup service, like writing a blog entry.http://www.iweek-interim.com/news/229208989?cid=RSSfeed_IWK_AuthorsAt least the T-shirts they're giving away at <a href="http://backthefup.net/">backthefup.net</a> do. As the site says, "Screw Klondike&#174; Bars, What Would You Do For A Back The F:\ Up T-Shirt?" Turns out what you have to do is something that promotes EMC's Mozy backup service, like writing a blog entry.To be more specific, you have to e-mail a proposal to Mozy, have them decide that, in their own words, your idea is awesome and doesn't suck, and then send your shipping information with proof you've followed through. <P> While "Back the f:\ up!" might be pushing the boundaries of good taste for some people, retail chain French Connection UK has gotten away with fcuk for years, so standards today must be pretty low. <P> This blog entry probably means I qualify for a free T-shirt, but somehow they never have my size (5X). So I'm putting together a video for its <a href="http://www.computernightmare.com">ComputerNightmare</a> site where I can win a Mac Pro or trip to Hawaii for me and Ms. Claus just for making the best video about a computer's demise. <P> Trebuchet? Thermite? Skipping a MacBook across the Hudson? So many options and only 48 days to make my video.2008-11-21T14:17:53ZIt's no surprise to readers of this here blog that the online backup market is hot. Even so, AOL has managed to fail at it and will be closing the pioneering Xdrive, founded in 1999 and acquired by AOL in 2005 for a reported $30 million. It will be shut down on Jan. 12, 2009.http://www.iweek-interim.com/news/229208763?cid=RSSfeed_IWK_AuthorsIt's no surprise to readers of this here blog that the online backup market is hot. Even so, AOL has managed to fail at it and will be closing the pioneering Xdrive, founded in 1999 and acquired by AOL in 2005 for a reported $30 million. It will be shut down on Jan. 12, 2009.I remember being an early user of Xdrive, which rather than providing a backup application installed software that mapped the X: drive of your Windows machine to their data center. Back in 1999 it was way cool, and as long as you didn't have lots of data to upload free. <P> Today Mozy, Carbonite, and a hundred or more others offer online backup and AOL is more interested in monetizing its assets via ad sales rather than subscription fees, according to an e-mail AOL exec Kevin Conroy sent out in July that TechCrunch posted <a href="http://www.techcrunch.com/2008/07/24/full-text-of-aol-email-xdrive-aol-pictures-mymobile-and-bluestring-to-shut-down/">here</a>. <P> Also to close are the BlueString and AOL Photo media sharing sites. <P> Competitor SpiderOak is offering 10% off to Xdrive users. <P> While I'm still an online backup fan, do-it-yourself and peer-to-peer systems like CrashPlan are looking like a better idea. I'm becoming a CrashPlan hero for my friends and family letting them backup to my systems. More on that another day.2008-11-20T17:15:59ZThe merger of Brocade, the clear leader in enterprise Fibre Channel switching, with Foundry Networks, one of the group with Extreme and Force10 that keeps Cisco honest at the high end of the Ethernet switching market, seemed like a good match back in August. With Cisco and the HBA vendors (Emulex and QLogic) pushing FCoE as the best thing since Fibre Channel itself (iSCSI's just for kids, you know) Brocade had to team up with an Ethernet switch vendor to try selling FC/FCoE switches that users plugged into Cisco Ethernet gear down the line.http://www.iweek-interim.com/news/229208757?cid=RSSfeed_IWK_AuthorsThe merger of Brocade, the clear leader in enterprise Fibre Channel switching, with Foundry Networks, one of the group with Extreme and Force10 that keeps Cisco honest at the high end of the Ethernet switching market, seemed like a good match back in August. With Cisco and the HBA vendors (Emulex and QLogic) pushing FCoE as the best thing since Fibre Channel itself (iSCSI's just for kids, you know) Brocade had to team up with an Ethernet switch vendor to try selling FC/FCoE switches that users plugged into Cisco Ethernet gear down the line.Like everything else in the world, acquisitions take credit. With the implosion of the credit markets this fall, the $3 billion deal has morphed at least twice into smaller and, at least for my friends at Foundry with stock options, less-profitable deals. <P> In October Brocade modified its offer from $18.50 plus a buck and a quarter's worth of stock to $16.50 a share, trimming $400 million off the deal. <P> Despite the $1.1 billion line of credit Brocade arranged to pay for the deal, it doesn't seem to be going smoothly. <P> This week's filings with the SEC revealed that Brocade gave Foundry permission to not only entertain, but also to solicit, other offers until tomorrow. We can only surmise that Brocade's management isn't happy with the financing options available to them. Where's Michael Milken when you really need him? <P> If the deal doesn't get done this year, Foundry can walk away and pocket $85 million or $125 million, depending on the details.2008-11-19T14:41:26ZRiding the coattails of Microsoft's announcement of its hosted Exchange service Exchange Online, Cemaphore Systems announced that its MailShadowX product will sync Exchange Online mailboxes with mailboxes on an organization's in-house Exchange server.http://www.iweek-interim.com/news/229208992?cid=RSSfeed_IWK_AuthorsRiding the coattails of Microsoft's announcement of its hosted Exchange service Exchange Online, Cemaphore Systems announced that its MailShadowX product will sync Exchange Online mailboxes with mailboxes on an organization's in-house Exchange server.The Exchange Online mailbox can then provide users with access to their mailbox contents, and the ability to send and receive mail from the same e-mail address they normally use for just $10/mailbox a month. <P> Setting up a comprehensive disaster recovery system for Exchange isn't for the faint of heart. Not only do you need a doppelganger server for your Exchange servers, and of course a DR site to house it, but also a process to make sure user's PCs running Outlook can reconnect to their mailboxes when you fail over. Between server hardware, replication and failover software from Double-Take Software or Neverfail, an exchange license for the doppelganger, etc., you're looking at a minimum of $8,000 on top of rent and bandwidth costs. <P> For organizations or locations with fewer than 50 users, the $50 for MailShadowX and $10/month for Exchange Online makes a lot more sense. This is especially true for branch offices, where the combination could make eliminating the local Exchange server a possibility. <P> In addition to mailbox contents, MailShadowX also syncs the Exchange global address list so users that primarily use Exchange Online will still have access to the corporate mailing lists and directory. <P> Like Cemaphore's MailShadowG, which syncs with Gmail, MailShadowX runs on the user's PC, but unlike the Google version doesn't run within Outlook. I'm looking forward to a version that can run as a service on a PC or server that handles sync for multiple users, which would really make eliminating the branch office Exchange servers worthwhile. <P> Cemaphore also announced general availability of MailShadowG and that MailShadowG could act as a conduit between Exchange and T-Mobile's Android phone powered by Google software. <P> In many years of installing, managing, and supporting Exchange servers, I've come to the conclusion that organizations without full-time IT staff are probably better off not trying to run Exchange themselves. It's hard to back up, hard to restore, and requires a lot more IT chops than Mike the PC guy probably has. If you're in this boat, fire up MailShadowX, turn off the Small Business Server's Exchange when all the mailboxes are synced, and just run Exchange Online.2008-11-11T12:49:25ZBetween the shift to disk backup and the economy rolling downhill, times are tough for tape library vendors. While IBM and Sun can shift their sales from tape libraries the size of a small Winnebago to their home-built VTL, the makers of midrange tape libraries are having a tougher time as much of their sales came through OEM deals with EMC, HP, or HDS and those vendors' VTLs don't pay Quantum or Overland's rent. Even media vendor Imation is hurting.http://www.iweek-interim.com/news/229208816?cid=RSSfeed_IWK_AuthorsBetween the shift to disk backup and the economy rolling downhill, times are tough for tape library vendors. While IBM and Sun can shift their sales from tape libraries the size of a small Winnebago to their home-built VTL, the makers of midrange tape libraries are having a tougher time as much of their sales came through OEM deals with EMC, HP, or HDS and those vendors' VTLs don't pay Quantum or Overland's rent. Even media vendor Imation is hurting.Quantum is being threatened with delisting by the New York Stock Exchange for trading at less than $1 a share for more than 30 consecutive trading days. While they can fix that problem with a simple reverse stock split, Quantum's fundamentals are nothing to write home about. The company that just 2 years ago bought ADIC for $770 million now has a market cap of just $57.6 million and an interest bill of $56 million a year that's in no small part still paying the ADIC bill. <P> On the plus side, Quantum has a respectable line of deduping appliances (but only sold $19M worth of them last quarter) and holds the Rocksoft patent on variable block size deduping. Software and patent license revenue from Dell, EMC, and Riverbed should start coming in soon, providing a new income stream. <P> Overland Storage is in even worse shape, losing $6.9 million last quarter with just $5.3 million in cash on hand. Their problems started when HP chose another vendor to supply midrrange tape libraries a while ago, cutting sharply into its income. <P> It also has VTLs, including a deduping model using the Diligent software, and recently picked up the Snap server line from Adaptec and is trying to make the transition to a disk-based economy, but management is desperately searching for $10 million to make it through 2009. <P> And even Imation is tightening its belt, laying off 200-plus employees (10% of the workforce) and closing a tape plant in Tucson, Ariz. <P> I'm thinking the ADIC guys timed this one just about right.