Server Virtualization

Description

• A "virtual server" is a software implementation of a server that executes programs like a real server. Virtualization is a method of running multiple independent servers on a single physical server. Instead of operating many servers at low utilization, virtualization combines the processing power onto fewer servers that operate at higher utilization. (See Figure 1 below.)
  
  Figure 1: Virtualization
• Virtualization can drastically reduce the number of servers in a data center, thus decreasing electricity consumption and waste heat, and consequently the size of the necessary cooling equipment. Some investment in software and hardware may be required to implement virtualization, but it is usually modest compared to the savings achieved.
• With today's servers, consolidation ratios in the 10:1 to 15:1 range can be achieved without placing stress on server resources.

Savings and Costs

Savings:

• Virtualization can reduce data center energy expenses by 10%–40%.¹
• Generally speaking, virtualization allows you to retire servers and/or defer purchases of new servers. Decreasing the number of physical servers does more than reduce the overall energy consumption of the servers; it also has a positive ripple effect through the entire data center. Server consolidation can result in reduced cooling load (the energy used to remove heat from the data center) and longer uptime when the data center is running on uninterruptable power supply (UPS) or generator power. In addition, having fewer servers requires fewer interconnects, which reduces IT configuration and maintenance costs. The result is improved service levels and greater energy efficiency across the data center.²
• Virtualization also improves scalability, reduces downtime, and enables faster deployments. In addition, it speeds up disaster recovery efforts because virtual servers can restart applications much more rapidly than physical servers. With virtualization, you can move entire systems in just a few seconds. You can also move a system from one physical server to another to optimize workloads, or you can move servers around for maintenance without causing downtime. Some virtualization solutions also have built-in resiliency features, such as high availability, load balancing and failover capabilities.
• BC Hydro estimated a payback of roughly 2 years based purely on IT energy savings.³
• VM Ware's return-on-investment calculator lists a default payback of 2 years on IT energy savings.⁴

Costs:

• Labor costs for:
  • conducting a thorough inventory and consulting all potentially interested parties,
  • migrating applications, and
  • removing old servers and deploying new servers (if necessary).
• New servers (if necessary). While virtualization can work with your existing hardware, you often achieve optimum results by implementing it with new, energy-efficient servers.
• Virtualization software. Note: some utility companies offer rebates for reduced power consumption, which could help offset the costs of virtualization projects. Talk to your utility representative during the planning phase to understand what financial incentives are available to you.
• End-of-life disposal costs for systems that are not repurposed.

Considerations

• Take inventory. The first step toward virtualization is discovering and identifying all servers in the organization. Take an inventory of their computing resources and their associated application workloads. It is easy to lose track of servers as time goes by, as mergers and acquisitions take place, as people procure servers without going through central IT, or as servers are retired or repurposed.
• Identify servers that should not be considered for virtualization. Servers that have privacy, security, or regulatory restrictions, or require ultra-high service levels, tend not to be the best candidates for virtualization or consolidation.
• Group remaining servers by workloads. Group the server inventory into pools of workloads that can coexist together. "Innovation" servers and "Production" servers would most likely not coexist. As previously mentioned, some servers may not be able to share resources due to technical, political, security, privacy, or regulatory concerns. Answering the following questions allows you to pool workloads into groups and determine the computing resources required for each group:
  • Who owns the server?
  • What is the purpose of the server?
  • What is the service level requirement?
  • What is the location of the server?
  • Is there a requirement for this workload to remain on a dedicated physical resource?
  For example, the exercise may yield a list of 100 Innovation workloads that all reside in the same location. You then can determine the resource requirements for the entire group. In this case, you may need 5 to 10 servers to support the resource requirements of these 100 workloads.
• Estimate the processor resources required for an application workload. The easiest method, one that is used by the vendors providing consolidation and analysis tools, is to simply multiply the reported utilization (% of processor utilized) by the total processor capacity (maximum processor frequency * number of processors * number of cores). This produces a MHz requirement that can be applied to target servers. You can refine this calculation by normalizing for multi-core overhead as well. Remember that the average processor is underutilized and typically is not the resource that limits consolidation.
• Keep in mind that the application workload can be depend on the time of day, week, month or longer if the business cycle dictates. For example, some applications are used primarily during business hours, while others peak during off hours. Servers should be sampled at least once per hour over a 30-day period to develop a complete workload profile. This technique ensures that peak and sustained-load times are captured in addition to normal usage.
• Combine workloads that have complementary characteristics. For example: Workload A consumes 10% processor capacity during business hours and very little at night. Workload B, which consumes 10% processor capacity during the night and very little during business hours. Combining the two workloads delivers the best possible results.
• Because virtualization generally saves power, it can affect how data centers are powered and cooled. For example, consolidating into a smaller footprint is likely to create "hot spots" in your data center. You will need to work with HVAC experts to ensure an adequate and efficient cooling configuration.
• There is always some risk of downtime. When migrating applications and files from an existing server to a new system, proper preparation can substantially mitigate the risk of downtime.
• Have a system for tracking all virtual machines (VMs). They are so easy to deploy, they can get out of control if you are not careful. If new VMs are created for specific projects, be sure they are not forgotten once those projects are completed.

Useful Links

• Virtualization to Improve Data Center Efficiency, The Green Grid, 2009