4 FM quick reads on data centers
1. Tracking PUE Can Help Create More Efficient Data Centers
This is Casey Laughman, managing editor of Building Operating Management magazine. Today's tip is that power usage effectiveness, or PUE, can be a good metric for measuring data center efficiency.
Calculated by dividing total site load by IT load in kilowatts (kW), PUE is a good gauge of a facility's energy efficiency, but relying on it alone can be misleading. For example, a new data center designed and equipped for energy efficiency and future expansion which is not yet operating at full design load initially will have a poor PUE. PUE will also degrade if an owner installs new servers with more energy efficient power supplies in an existing data center.
But, as an overall snapshot of how efficient your data center is, PUE is effective. The goal of the "PUE Arms Race" is to drive down power usage effectiveness to 1.0, where the only energy used is the energy powering the computer.
There are three fundamental ways to improve the energy efficiency of a data center. One way is to install new computer equipment with more efficient power supplies, and this is often done as owners periodically refresh their computing equipment. Another is to implement on-site power generation, for example, cogeneration or solar power. These grand-scheme approaches are not often implemented today, but they have increasing potential as the technologies improve and their capital cost decreases.
The third approach is to design, engineer and operate data centers to maximize the efficiency of the building infrastructure. Whatever else an organization is doing, this is fundamental to improving energy efficiency. Here is a look at some of the leading trends in mechanical and electrical systems. Many of the techniques being implemented by data center owners and engineers have an established track record in non-critical facilities. With data center energy costs escalating, these techniques are making their way into the mission-critical arena.
Consider the Cloud When Expanding Data Centers
This is Casey Laughman, managing editor of Building Operating Management magazine. Today's tip is that cloud services are becoming a more common option for data center expansion.
The major advantage of expanding to the cloud is that the organization pays only for access to the services it needs to meet its business objectives, not for ownership of capital assets. The approach offers an immediate, scalable solution for a monthly or annual fee.
Although cloud services are touted as a new concept, they have actually been available for more than 50 years under different terms, including time sharing and partitioning. Today, cloud services take three major forms: software as a service (SaaS), platform as a service (PaaS), and infrastructure as a service (IaaS): SaaS is the most popular form of cloud services. A service provider offers software to support the end user's business. The end user can configure the software to suit their needs, although they cannot change or modify it.
PaaS offers a platform to clients for various purposes. For example, Microsoft Windows Azure offers a platform for developers to build, test and host applications that can be accessed by the end users.
IaaS offers infrastructure on demand ranging from storage servers to applications to operating systems. For example, Microsoft Office 365 provides applications and storage space. IaaS enables an organization to save on the capital costs, space and staff it takes to set up and maintain in-house infrastructure.
While cloud computing can be an option for data center expansion, remember that the decision will hinge on business analysis. Apply core financial analysis and weigh the payback period, net present value and internal rate of return of cloud computing versus other expansion options. At the end of the day, the technology solution must support the business.
Two Fundamental Ways to Increase On-Site Capacity
There are two fundamental ways to increase on-site capacity: power density or square footage. Typically, owners and managers consider expanding the physical capacity of a data center when they can no longer increase power density to support the business operations.
As a prerequisite for expansion, the organization must have the appropriate management processes, procedures and oversight in place to enable a seamless expansion of critical infrastructure while maintaining operations and protecting data throughout the project. Before the project begins, the entire process must be documented, including a rigorous quality review procedure for data center operators to execute transitions without outages.
Companies that have successfully completed an on-site expansion in the past may be able to leverage existing procedures for a new project, but they typically lack adequate internal resources to manage it. As a result, they must turn to experts when they are expanding a live data center. In addition to managing the risks to data and operations, the owner and project team must manage real safety risks to operating personnel during "hot" work.
There are several advantages to existing site expansion. The biggest is maintaining control over mission-critical assets, especially their performance and security. Existing site expansion also avoids the inefficiencies and risks of compartmentalizing applications and outsourcing services. The organization gains economies of scale by leveraging existing real estate and operational resources, including staff and management. Growing in place also avoids the costs associated with decentralizing operations.
The major business risk associated with existing site expansion is the tendency to over-project demand and overbuild the facility. There are several reasons for this. A build-out typically requires 12 to 18 months to complete. In anticipation of this lengthy build process, an organization may try to project demand three or more years into the future. However, the longer-term the business forecast, the more likely it is to be inaccurate. Even companies that are very skilled at demand management can experience spikes, for example, due to a new product with an unexpectedly high demand. So there is a tendency to overbuild "just in case" unforeseen circumstances arise.
The carpenter's mantra, "measure twice, cut once," applies to all data center expansions, but particularly when considering an on-site physical expansion. Before moving ahead with an expansion, benchmark the existing data center's baseline performance to identify stranded capacity and opportunities to improve the facility's operating efficiency. For example, some owners have gained 30 percent or more capacity by implementing cost-effective changes in the management of air flow. If an expansion is warranted, the baseline performance benchmark will be a means of assessing the effectiveness of the project.