Mar 21st, 2011 |
Interop Las Vegas 2011’s Data Center Track
When you begin your next data center refresh cycle, you are going to have more choices to make about architectural direction than in previous years where the decisions were primarily around price and performance. Granted, price and performance are important, but increasingly building silos of applications is less effective. We have certainly seen that, in many cases, server virtualization results in fewer servers needed when you can combine under-utilized virtual machines onto a single platform. We have seen a decrease in recovery point objectives considering you can move a VM in a fraction of the time it would take to install a new server and restore the application from a back-up. And we have seen a reduction in downtime because we can move running virtual machines without any impact on application performance.
In the coming years, you can expect to reap similar benefits in network and storage virtualization that you reaped with server virtualization. Simply virtualizing data and storage networking can reduce cabling costs, increase capacity, and reduce overall network utilization by letting traffic take shorter paths. Many of those benefits are tied directly to lossless Ethernet, which Andre Kindness of Forrester Research covers in Understanding Data Center Bridging where you will learn how the DCB protocols will impact data center network design and operations.
Data center networks will also be replacing Spanning Tree with real multi-path Ethernet technologies. Multi-path Ethernet is going to be a big topic in any data center design and the choices vendors are making will impact your purchasing and design decisions. We have a stellar panel of experts including Brook Reams from Brocade, Francois Tallet from Cisco, and Jeff Kabul from HP on Making Sense of Multi-Path Ethernet Networks who have experience designing and building large scale networks and will be able to explain—at a practical level—the benefits and drawbacks with multi-path Ethernet. Bring your questions and your popcorn.
The changes in the data center will also mean changes in IT operations. Chances are you already use some automation in your data center operations and you can expect that to expand. However, the real benefits come with orchestration—integrating and automating multiple data center management processes, a topic that HP’s Terry Gates will be diving into with Managing the Migration from Automation to Orchestration on Wednesday. The farther data center operations get into automation and orchestration, the closer they will need to work with developers to streamline management and applications. The management strategy called DevOps, is the subject of Thursday’s panel DevOps and You where key industry experts Shlomo Swidler, George Reese, and John Willis, will explain what DevOps is and how it benefits your IT organization.
Finally, we want to keep things running smoothly. That is where Dave Peters is going to discuss System Architecture Design and Capacity Planning . Planning for capacity increases ensures that you can plan for purchases in advance and make better use of your existing resources. And with all this new equipment, the cost of power and cooling will certainly rise if you don’t get control of it. While it would be great to build a new data center with the latest green technologies, there are new technologies that can be applied to existing data centers that can reduce power consumption. Randy J. Ortiz from Internap and Jeff Rose from Vantage Data Centers, will be speaking on their experience in Maximizing Data Center Efficiency.
Hope to see you all at Interop Las Vegas in May.
