September 2012

Server Consolidation

Server Consolidation Checklist

By: Todd Bone, XS International

It’s no surprise; the economy has been in a downward spiral for some time now. In fact, nearly every industry has been affected by it to some extent. But, for corporate data centers there’s no doubt that the implications of cost and efficiency will weigh heavily long after the current economic downturn ends. So, what’s the best way to consolidate servers and avoid unwanted wasted resources?

Here’s a checklist for server consolidation strategies that work:

To Be or Not To Be Physical???
Server virtualization has emerged and brought to life a whole new cost effective approach to server storage. Virtual servers convert each application and operating system to a unique virtual instance called a virtual machine (VM). The use of VMs disconnect each instance from the underlying server hardware, drivers, and other elements that traditionally tied hardware and software together. The use of VMs offers data centers greater flexibility and control over their applications while reducing the number of physical servers needed. Simply there are fewer servers, and each virtualized server can perform more business work per dollar spent on hardware, power, cooling, management and maintenance.

Utilize Servers to Their Fullest Potential.
Don’t be fooled, that doesn’t mean your servers should be running at 100% capacity. Most traditional servers are only utilizing between 5% and 10% of their total computing capacity – a definite waste! Although it seems counterintuitive, data centers should never seek 100% utilization on virtual servers either. Data centers should leave an ample amount of processing capacity available to assume the processing burden of other machines. Why? Just consider what would happen if a data center had two servers, each running multiple VM’s and using close to 100% of server capacity. If one failed there would be no processing capacity left on the other server. The affected applications would have to remain offline until the server was restored. Ideally, a data center would want to run all servers at 50% capacity leaving ample room for failover protection. The key factor for data centers to consider is to balance consolidation with performance and failover protection.

Sometimes Three’s a Crowd
It’s important to make sure that all your servers and tools work together. In order to be cost effective data centers want to get more out of their server hardware. This often requires knowledge of the server hardware, the way the hardware is utilized by each VM, and the operating system hypervisor. There are many performance-monitoring tools that can track and report on the computing resources needed by each VM. There are several third party monitoring products as well as the major hypervisor vendor tools that can be used. The most important thing when selecting tools and servers is to make sure they are all interoperable. This is an effective way to ensure that one tool can monitor and report on all of the available servers which is more cost-effective than using multiple different tools from various venders.

Don’t Get Overtaken By a Virtual World
No, I’m not talking about Minecraft for all you gamers out there; I’m talking about Virtual Machines. Like Minecraft they are easy to create, but time consuming to maintain and keep under control. While nearly anyone can easily create a VM with a few clicks of the mouse, it’s pertinent to limit those who have access to create them. The more VMs the more management overhead and security risks come into play, ultimately reducing the cost-effectiveness of virtualization in the first place. The best practice is to define what does and does not constitute the need for creating a new VM, and limit the number of personnel that can create and manage them.

Out With the Old In With the New
From time to time data center servers will need repairs. A new memory card here, or CPU there is likely in the life cycle of a server. Upgrades and replacements are easy to perform, but they’re also an easy way to waste money. The trick is to determine the financial viability of each upgrade. When a system is new, it’s definitely in the data centers best interest to perform upgrades or buy replacement parts. As the system ages it becomes more important to evaluate the cost-effectiveness of IT replacement hardware verses scrapping the machine and buying a new one. Many companies prefer to buy refurbished IT hardware as a cost effective approach. If you do determine that it’s time to get rid of the old machine and buy a new one, there are companies that specialize in IT asset disposition and IT recycling that can help you find a cost-effective approach.

Server consolidation allows data centers to maintain and run their IT hardware in a cost-effective manner. With proper server consolidation planning you can increase your physical server capacity by as much as 80% by creating virtual machines. Through proper management data centers can create, maintain, and manage their virtual machines in a way that will reduce overall costs and effectiveness.

About XS International
XSi ( specializes in new and used IT hardware and software sales as well as data center consolidation and IT asset disposition.

Find out how you responsibly recover 5,600 square feet of Data Center floor space in less than a week with no disruptions to operations or production: