This audio was created using Microsoft Azure Speech Services
Data center service providers are naturally in tune with the many benefits of virtualization technology and how it can dramatically increase server utilization rates, thus reducing the number of physical servers required. Most likely, your customers are on board with the concept, too.
But as your data centers become increasingly virtualized, you’ll also need to take steps to ensure the supporting infrastructure is updated accordingly. Or, if you’re building a new data center that will support lots of – or even solely – virtual servers, you’ll need to plan accordingly.
The effects virtualization has on data center infrastructure is spelled out nicely in Schneider Electric white paper number 118, “Virtualization and Cloud Computing: Optimized Power, Cooling, and Management Maximizes Benefits” and in some previous blog posts on the topic, including this one.
In a nutshell, virtualization raises four issues:
- Increased rack density because more powerful servers are running at higher utilization rates
- Potential for increased power usage effectiveness (PUE) if power and cooling are not right-sized for new, virtual loads
- Dynamic IT loads that can vary in both time and location as they shift from one server or cluster to another
- The potential for reduced redundancy requirements in physical infrastructure due to increased fault tolerance in the IT infrastructure
The white paper goes into depth on each of them so I won’t repeat it here. But, as noted in a webinar I co-hosted titled “The Data Center Revolution: Simple is Better,” there is a tool that can alert data center operators when conditions are getting out of whack due to issues including virtualization: data center infrastructure management (DCIM) software.
DCIM software enables data center operators to aggregate data from disparate systems such that they can get an integrated view of the physical building infrastructure and IT equipment. In the virtualization example, that means they’ll be alerted to any hot spots cause by racks that are too dense for the cooling infrastructure that supports it.
DCIM tools also enable operators to conduct “what if” scenarios to help correct such problems. You can see the effect of moving a rack or server cluster to a different area of the data center, or find the best spot to place a new cluster or rack from a power and cooling perspective.
The tools come with simple, web-enabled dashboards that give easy access to key performance indicators such as power, cooling, network and PUE. Schneider Electric also offers applications that provide access to all this data from mobile devices, so you can be alerted to issues and address them even when you’re far away from the data center in question.
