Subscribe to the RSS feed

As we described in the last few entries, adding storage software to standard servers or even to virtual servers is making more sense for a variety of data centers because of the increased performance capabilities of the Intel processor family. Despite this there may still be times in certain environments where there is a need for specialized storage hardware.

Comment on this blog entry

Cloud storage is constantly being discussed in the IT media today. When you get right down to it, what can businesses really use cloud storage for now? The small office, individual user has embraced cloud storage for backups and for collaboration, but what can larger businesses use these services for?

Comments(2)

In our last few entries we looked at what can be done today with storage software running as virtual machines. In this entry we will consider what the future holds for storage as a virtual machine. Storage as a virtual machine may be the only way you apply data services in the future.

Comment on this blog entry

Completing our storage as a virtual machine re-interviews were conversations we had with EMC and Nexenta. While our last entry focused on systems that leveraged virtual machines to deliver block I/O storage services these two companies are delivering something a little different, NAS services and backup services.

Comment on this blog entry

There are countless ways to manage data available to the storage manager today but most of these solutions look at data as a problem. Few take an asset view of data, understanding that it is something to be cultivated and leveraged for future use. Storage managers should do more than just manage their data, they should know it.

Comment on this blog entry

As we dive deeper into the storage as a virtual machine concept we went back and re-interviewed some of the players in the storage as a virtual machine market, focusing specifically on what they provide. The first two conversations were with DataCore and HP. We will cover more suppliers as the series unfolds.

Comment on this blog entry

Data storage capacity is cheap. For most environments obtaining enough capacity is no longer a challenge, it is managing that capacity that becomes the problem. Growth, especially in unstructured data, continues unabated. Deciding what data should be where is one of the biggest challenges that the storage manager has to face today. Users don’t want to think about where data should be stored and storage managers don’t have the time to think about it.

Comment on this blog entry

Traditionally storage systems and other storage related services have been delivered as customized systems. This was done to maintain performance and to reduce support costs to the manufacturers. As server technology continues to increase in performance, the concept of providing storage services as a standalone application installed on your own server hardware is becoming increasingly popular. Now with virtualization the storage as an application concept is being applied to virtual machines.

Comment on this blog entry

I recently concluded a series that examined the components of the storage environment that can impact overall storage I/O performance. There was storage I/O bandwidth, controllers and drives. What if you are like many data centers and you don’t need to wring out every drop of storage I/O performance from your storage infrastructure? What should you do with too much storage performance?

Comments(1)

Most of the major Solid State Disk (SSD) manufacturers and providers are reporting record sales both in terms of units and capacity being purchased. Much of this success is being driven by cost reductions in the technology and an increased understanding of how to best implement the technology. Mainstreaming SSD is going to require more than just price drops, its going to require intelligent leveraging of the technology.

Comments(1)

In our last entry we talked about the use of cloud storage as a backup target, but another ideal use case for cloud storage is to use it as an archive area. Almost every IT organization has old data that they want or must keep, but are struggling with where to keep it. Its ability to identify, automatically move and transparently recall data could make file virtualization the ultimate cloud gateway.

Comment on this blog entry

Cloud based backup services have been successful in the consumer space. Companies like Mozy, Carbonite and others are protecting thousands of laptops and home desktops, but can cloud based backups services move beyond protecting consumer or prosumer data and into the data center? Are cloud based backups ready for business?

Comment on this blog entry

The storage component of a virtualized server infrastructure has been labeled as complex and expensive. In our prior entries about selecting a storage foundation we discussed what systems and protocols are available that might help simplify and reduce costs for storage in a virtualized environment. Beyond physical hardware you need software tools that can link the abstract virtual machine to the physical storage.

Comments(1)

In the last performance entries we discussed understanding storage bandwidth and understanding storage controllers. Next up is to understand the performance characteristics of the hard drive itself and how the mechanical hard drive can be the performance bottleneck.

Comment on this blog entry

Whether your company's data center is a couple of servers stashed in a closet or a gleaming, state-of-the-art climate-controlled facility, you're still facing the same set of challenges: how to keep the IT lights on while controlling costs, take advantage of new technologies to stay competitive, and position your company for an economic recovery in the midst of the toughest times for IT that many of us can remember.

On Wednesday, October 21, 2009, help is on the way.

Comment on this blog entry

At Storage Networking World, at least one new category in storage is coming to the forefront; Automated Storage Tiering. These are typically devices that can sit in front of your existing storage platform and allow some of it to leverage a high speed solid state front end without you manually having to move data to a Solid State Disk (SSD).

Comment on this blog entry

Storage controllers are the engine that drives the storage system you own. They are essentially a compute engine for storage arrays. Understanding storage controller performance and what can impact storage controllers is an important step in the optimization of your storage environment. It is also something that many storage managers assume is good enough.

Comment on this blog entry

HDS announces a move into public and private cloud storage, primarily through rebranding

Comment on this blog entry

Storage bandwidth is the connectivity between servers and the storage they are attached to. When it comes to understanding storage bandwidth performance you have two challenges to deal with. The first and most obvious is can the storage get the data to the application or user fast enough? The second and less obvious is can the applications and hardware those applications run on take advantage of that bandwidth?

Comment on this blog entry

For most storage managers improving storage performance is an endless loop of upgrades that are taken until the problem goes away. Understanding where to look and how to configure the environment is often a series of "best guesses" instead of a thorough understanding of it. In today's economy best guesses are not allowed. Making the right move, the first time, is critical.

Comment on this blog entry

In our last entry we discussed selecting the right storage foundation and I advised that you may want to initially ignore what protocol to use. That said, part of building a storage foundation for server virtualization is selecting the protocol.

Comment on this blog entry

The storage component of a virtualized server infrastructure has been labeled as complex. The storage and server virtualization suppliers have both tried to deliver solutions that reduce storage complexity in server virtualization projects. The challenge for virtual infrastructure administrators is that there are so many options that it can be confusing. There are several steps to take when selecting a storage foundation for virtualized servers and our next series of entries will cover these steps.

Comment on this blog entry

Storage scalability in a virtualized environment is quickly becoming a concern for some data centers. Unlike the very predictable single server world which had a single application, single NIC card and single host bus adapter, the virtualized host can have dozens of virtual machines and multiple network interface cards. This leads to a very unpredictable and random workload that can push storage controllers to their limits.

Comment on this blog entry

One of the most expensive components of a server virtualization project is the storage which it is attached to. This is one of the reasons that VMworld seems more like a storage event than a server virtualization event. One of the key themes from vendors at this years event was reducing the storage costs in server virtualization projects.

Comments(2)

Most green storage methods really allow you to store more data in the same physical space, for the same amount of power consumption. To get serious about power efficiency you have to be able to turn things off. The ideal way to do this is have the drives either spin down or turn them off, but there is limited information about implementing power efficient drives.

Comment on this blog entry

There is an ever increasing emphasis by storage suppliers on the power efficiency of their systems. There is also an increasing interest from IT managers on making their environment more power efficient. This is being driven not so much by the desire to be environmentally sensitive, but more because an increasing number of data centers simply can't get more power to the building.

Comment on this blog entry

One thing suppliers and analysts are quick to point out is that when it comes to data protection it is not about how well you backup, it is about how well you recover. That sounds very catchy and for the most part is accurate. I believe however, that backup is an equally important part of the data protection puzzle. It is after all poor backup strategies that make recovery so hard and unpredictable.

Comment on this blog entry

Solid State Disk (SSD) is often the solution to drive up performance of a particular application, increasing response time to users and thereby increasing revenue or productivity. There can also be significant CAPEX savings by implementing SSD, what is often missing from the discussion is the operational or OPEX savings that comes from implementing SSDs.

Comment on this blog entry

Traditional volumes have been around as long as we have had file systems, but now may be the time to pronounce the end of the storage volume. As data centers get larger and storage systems are populated with more and more physical drives they are growing into a storage manager's nightmare. There are just too many volumes of drives to be managed effectively.

Comments(1)

In my last entry we discussed some of the challenges storage presents to expanding the level at which companies deploy virtualized servers. There is another, potentially larger roadblock to virtualization; staff inefficiency.

Comment on this blog entry

After the initial rollout, server virtualization projects often get stuck as managers and administrators begin to cope with how successful the first phase of the project was and begin to try to grasp how they can increase the number of systems that are virtualized. There are two big roadblocks to a high percentage of virtualization in most data centers; cost and efficiency.

Comments(1)

At VMworld this year there is a lot of emphasis on making the storage easier and less expensive for the small to medium sized businesses that are taking their first steps into the server virtualization market. The focus of these storage vendors is on driving down costs and driving out complexity.

Comment on this blog entry

In my last entry we discussed Making Data an Asset. This entry will focus on where that data asset should be stored. What is needed is a strong storage foundation, one that is designed to last for years, if not decades, but also one that will store that data efficiently and of course be complimentary to the enterprise class indexing that we described in our last entry.

Comment on this blog entry

Data is often looked at as a liability; something that has to be stored, protected and preserved. Data storage has led to massively expanding storage environments and such initiatives as archive. Protection has led to incredibly elaborate backup and recovery schemes and preservation has led to eDiscovery and compliance. All of these processes are reactive, how can the view of data be changed to proactive, to using data as an asset?

Comments(1)

One of the storage management challenges we see every day in customer data centers is there are too many copies of data in circulation. Ironically its this fact that built much of the value and motivation behind data deduplication. It should not be this way. Why should you get to a last copy of data?

Comments(1)

The challenge that most archive systems have is they are too big for the job. Some organizations, especially in the small to medium sized business market, may not want or need to move all their inactive data to a secondary storage tier, yet they know they have specific electronic documents that from time to time need to be retained and locked down.

Comments(1)

There is a dizzying array of storage solutions available to storage managers today. Whether its backup, archive or primary storage there are multiple options available. Many times manufacturers try to position themselves as a single source of storage solutions for a data center. Be careful of this approach, seldom is one manufacturer able to provide best of breed solutions in every product category.

Comment on this blog entry

I recently read a claim by one major supplier of Fiber Channel over Ethernet (FCoE) technology that it would be the dominant infrastructure in use in data centers in two to three years! Are you kidding me? Other than impossible that is just not the speed at which the data center moves. The data center evolves, it does not revolt.

Comment on this blog entry

With all the products that are available today for optimizing storage through deduplication and/or compression, one of the best methods available is deletion and reclamation.

Comments(1)

In a recent series of entries I covered several storage technologies that can help a data center maximize their CAPEX. Most of that series focused on cutting costs by using less primary storage either through archiving or efficiency. Another way to maximize your CAPEX investment is to maximize IOPS with SSD (Solid State Disk) technology.

Comment on this blog entry

Cloud storage is ideal for small businesses to perform backups, archive important data and to share files when collaborating on a project. Small businesses are typically built on desktops and laptops making the cloud the ideal connection point. As the business grows and servers are added, applications become more critical and the cloud becomes the bottleneck. At this point to handle the needs of a growing business cloud storage needs a hybrid.

Comments(5)

With the cost of direct attached storage (DAS) dropping and the capacity that it can deliver for those dollars increasing, you would think that the demand for shared storage is dwindling. Reality is that shared storage is on the rise and the biggest reason for its growth has little to do with storage management or even data protection. Those are nice side benefits, however the real motivator is server virtualization.

Comments(1)

Egnyte, which blends local and online storage for SMBs, pulled in millions in VC investment today.

Comment on this blog entry

We've spent the last several entries discussing the unification of storage and there is one aspect of unification that I have not discussed; unifying the infrastructure. I do currently have a series of videos currently running with Information Week on FCoE so in this entry I'll just touch on the high points.

Comments(1)

The last few entries we have been covering aspects of unified storage. The bottom line is that unified storage has it’s place and many organizations can benefit by having these systems, but where does this leave storage platforms that essentially do one thing and do it well?

Comments(1)

The focus of storage unification has for the most part been protocols. Leveraging a NAS to also serve up a SAN protocol (fibre or iSCSI) and as I discussed in my last entry there is definitely a place for that. The forgotten part of storage unification however is unifying all the storage that already exists in the data center; this is typically done through a form of storage virtualization.

Comments(2)

As discussed in our last entry, unified storage is all the rage right now in the storage industry; it is essentiality the combining of NAS with a block protocol, like iSCSI or traditional fibre channel. The question is however do you need unified storage?

Comments(6)

What started as a whisper has now become a roar. All of a sudden every storage vendor you talk to has Unified Storage and all of a sudden you MUST have it. All of which begs the question, what is unified storage and do you need it?

Comments(3)

Solid State Disk is a mature, stable technology poised for widespread adoption in enterprises of all sizes. It solves performance and power issues that mechanical drives can not. Most data center managers, large and small, have an eye on this technology but are not exactly sure where to start with SSD.

Comment on this blog entry

Legal hold is a term used to set aside certain data to make sure it is not altered while a legal case is being settled. One of those situations is employee termination. The chances are there for the employee to file a wrongful termination lawsuit and for the data center that means placing exiting employees' data on legal hold.

Comment on this blog entry