Cloud // Infrastructure as a Service
Commentary
12/17/2012
02:08 PM
Charles Babcock
Charles Babcock
Commentary
Connect Directly
Twitter
RSS
E-Mail
50%
50%

Cloud Computing: Best And Worst News Of 2012

What were the key developments for enterprise cloud computing this year? Let's look at four big wins -- and three setbacks.

Win #1: Customers Have More Providers From Which To Choose.

Amazon, Rackspace and others have new rivals. Providers of co-location data centers and hosted services or managed services have found that the distance between what they have traditionally done and cloud computing, where customers do much of the management themselves and pay per use, is short.

SoftLayer Technologies added 30,000 servers to the 70,000 already in its 13 data centers to increase its IaaS offering. Smaller providers with a strong ties to their customers have emerged including Hosting.com, Bluelock and Peak 10, to name a few. Bluelock is one of many regional providers of IaaS whose appearance was spurred by the prevalence of VMware virtualization in the data center and the availability of VMware-compatible cloud management software for service providers. VMware officials in February said they had 94 such partners. Joyent with its SmartOS open-source Solaris IaaS, running in Joyent data centers or installed in customers' data centers, is another alternative, with powerful funding backers.

There's also growing competition in the software used to manage infrastructure clouds. The OpenStack open source code project has assembled seven cloud computing modules governing servers, storage, image management, networking and operational reporting. Rackspace, Dell, and the SUSE Linux unit of Attachmate now offer OpenStack-based cloud services, with software vendors Nimbula, Piston and Nebula offering wares to build a private OpenStack cloud

Win #2: Price Wars Bring Down Cloud Storage Costs.

This was the best news of 2012, and it came toward the end of the year with genuine competition in cloud storage. In theory, cloud prices should trend downward because the power of devices used in storage and computing have been increasing quickly, but it had been hard to see the theory in action -- until Dec. 1. That's when Amazon cut Simple Storage Service (S3) prices by 24% to 27%. The next day Google cut storage prices by 20% to remain in step. It was the biggest drop in cloud storage prices since the services were created.

Savvis is among the rival cloud vendors wrestling with how much to compete on price for IaaS. Before the latest plunge, Savvis had been planning to match Google and Amazon on storage prices while offering more value-added features, such as checkbox offsite data replication, said director of storage product management P. J. Farmer. Such an option would provide protection against loss of data or systems in a natural disaster, such as Hurricane Sandy. "Now I'll have to think it over," said Farmer, about matching those now-lower rates. Savvis' special features might let it hold prices a bit above Amazon and Google. But Farmer's dilemma shows the trend is both hard to escape and far from over.

Win #3: Telecom Provides Reliable Backup.

Why doesn't one cloud data center serve as a backup facility for another? That way, a natural disaster that robs one site of its ability to operate won't necessarily harm the businesses of its customers, who failover to the backup center.

The concept is simple, but the execution is hard. Duplicate systems must be stored in each location. And at the moment of crisis, it's impossible to transfer all the data needed by the system in a few minutes. So preparations need to be made. Data must be replicated to the backup site on an ongoing basis, so in the advent of a disaster, only a few days or few hours of data must follow the migration to the backup site.

With the launch of cloud services from data centers owned by telecom providers, however, the process got a little easier. Savvis, owned by CenturyLink, the third largest U.S. telecom supplier, has tied its cloud data centers together with CenturyLink private lines.

That creates a potential secure and reliable failover chain, if you should ever find you're on the East Coast with a hurricane bearing down, or on the West Coast after an earthquake.

It's a check-off option to replicate data created in one Savvis cloud center in another, provided you're already a CenturyLink telecom customer. In that case, the service is free over existing lines. Even if you're not, the possibilities of the new service are obvious: disaster recovery mechanisms that, if necessary, could be adapted on the fly by those with foresight and preparation. Savvis operates five data centers around the world where the service is an option.

Such options are likely to become more prevalent as the telecom providers weigh into cloud computing. Terremark, owned by Verizon, is a candidate to combine the strength of extensive private networking with cloud computing.

InformationWeek predicted such chains would emerge back in September 2011.

Win #4: More Cloud Server Types Available.

Amazon has led the way in setting the pace for types of virtual servers, offering micro, small, medium and large. It added extra-large and double extra-large to the list, then high-memory extra-large, double extra-large, and quadruple extra-large. Concentrated CPUs arrived in the form of high-CPU medium and high-CPU extra-large. Cluster instances, graphics processing unit instances, and high I/O instances round out the list.

Each vendor is doing some variation of this service catalog. But there is a further development on the horizon. Instead of the cloud telling you how much memory and storage you'll get with your chosen server size, you will soon be able to tell the cloud. Being able to configure exactly the server you want, with appropriate I/O and networking as well other components will be the final phase of giving customers choice in self-provisioning their virtual servers.

Previous
3 of 3
Next
Comment  | 
Print  | 
More Insights
Comments
Oldest First  |  Newest First  |  Threaded View
John Foley
50%
50%
John Foley,
User Rank: Apprentice
12/19/2012 | 5:04:19 PM
re: Cloud Computing: Best And Worst News Of 2012
Cloud reliability (or lack thereof) and cost are well-known issues, but the threat of VM snooping, a.k.a. side vector attack, is less well understood and in some respects more worrisome. So far, it seems to be a theoretical threat, but research now shows it can be done, and when there's a will, there's a way. The researcher concludes that "highly sensitive workloads should not be placed in a public cloud." Let's see how long it is before some company pays the price for failing to heed that warning.
TechYogJosh
50%
50%
TechYogJosh,
User Rank: Apprentice
12/20/2012 | 9:51:48 AM
re: Cloud Computing: Best And Worst News Of 2012
Charles, did you included only cloud infrastructure analysis by design? Or have you also converted to the mad rush which calls only IaaS as cloud and others as SaaS, PaaS, etc? In fact each of these SaaS PaaS IaaS are part of cloud metaphor and you should have touched upon at least all of these cloud layers. None the less a pretty good read.
Laurianne
50%
50%
Laurianne,
User Rank: Author
12/20/2012 | 2:35:29 PM
re: Cloud Computing: Best And Worst News Of 2012
While the security worries get more press, the almost-impossible-to-compare pricing issue is just as painful for IT managers day to day, although Amazon is starting to take positive steps. I wonder how long it will take to get to more apples-to-apples comparisons. The situation works in the vendor's favor right now, not in IT's.

Laurianne McLaughlin
InformationWeek
hbaldwin940
50%
50%
hbaldwin940,
User Rank: Apprentice
12/20/2012 | 4:50:31 PM
re: Cloud Computing: Best And Worst News Of 2012
Charles is dead-on with setback #3 (Gǣpricing is still a messGǥ). There are, of course, ways to get rapid fixed-price, fixed-schedule cloud deployments, as this SAP video shows: http://bit.ly/SWnnp4. About me: http://bit.ly/UoILDg.
Multicloud Infrastructure & Application Management
Multicloud Infrastructure & Application Management
Enterprise cloud adoption has evolved to the point where hybrid public/private cloud designs and use of multiple providers is common. Who among us has mastered provisioning resources in different clouds; allocating the right resources to each application; assigning applications to the "best" cloud provider based on performance or reliability requirements.
Register for InformationWeek Newsletters
White Papers
Current Issue
InformationWeek Tech Digest, Dec. 9, 2014
Apps will make or break the tablet as a work device, but don't shortchange critical factors related to hardware, security, peripherals, and integration.
Video
Slideshows
Twitter Feed
InformationWeek Radio
Archived InformationWeek Radio
Join us for a roundup of the top stories on InformationWeek.com for the week of December 14, 2014. Be here for the show and for the incredible Friday Afternoon Conversation that runs beside the program.
Sponsored Live Streaming Video
Everything You've Been Told About Mobility Is Wrong
Attend this video symposium with Sean Wisdom, Global Director of Mobility Solutions, and learn about how you can harness powerful new products to mobilize your business potential.