Here’s a roundup of some interesting items we came across this week in our reading of data center industry blogs.
The Data Center Outsourcing Decision Process Needs Serious Rethinking - Mark Thiele at Switch examines the changing criteria for picking colo: "The nature of how you acquire and use IT has changed, and continues to change to a model more aligned with buying 'services.' Even though you may still decide to build and operate your own infrastructure the way you add or remove capabilities will be (should be) much more Lego like (See Fluid IT). My suggestion on the IT environment of the future is that the flexibility and speed of service change everyone will drive towards, will force us to rethink how we buy. It will also force us to seriously consider 'where' we buy. No, I don’t mean HP vs. Amax or Cisco vs. Juniper, it’s more of a 'location' where."
Top Problems with the TOP500 - Where was Blue Waters in the recent Top 500? The new NCSA system didn't field an entry, and NCSA's Bill Waters explains why: " the TOP500 does not provide comprehensive insight for the achievable sustained performance of real applications on any system—most compute-intense applications today stress many features of the computing system and data-intensive applications require an entirely different set of metrics. Yet, many sites feel compelled to submit results to the TOP500 list 'because everyone has to.' It is time to ask, 'Does everyone have to?' and more specifically, 'Why does the HPC community let itself be led by a misleading metric?'"
Cloud-native, cloud-centric, or cloud-ready? - While I'm weary of efforts to define and parse cloud terminology, the latest jargon is discussed in a thread convened by CloudScaling's Randy Bias: "I am trying to understand if we are close to a consensus on the new apps driving all of the cloud growth. Cloud-native is a term positioned by the folks at MessageBus, cloud-ready is what we at Cloudscaling have been using for a while, and cloud-centric is an IBM-ism."
When an HP cloud is not an HP cloud (and whether it matters) - At GigaOm, Barb Darrow looks at the rise of third-party clouds: "Much of Hewlett-Packard’s new cloud relies on non-HP technology. The foundation is OpenStack; the content delivery network (CDN) is from Akamai; and the latest piece, unveiled this week, is a Platform as a Service from ActiveState. That outsourcing of key technologies from a company once known for its “invent” motto, is jarring to some. Others think it’s weird given the tens of billions HP spent on technology acquisitions in the past few years. With all that tech in-house, why send out for more?"
Why Virtual Machine Recovery Is No Piece of Cake - Madhu Reddy discusses VM disaster recovery at the SunGard Availability blog: "Some vendors like to tout how easy DR becomes with virtualization. Sure, provisioning, deploying, and even moving virtual machines (VMs) to new servers are relatively simple operations. But recovering VMs in large-scale virtual applications environments presents several protection and recovery challenges."
