Archive for April, 2010

Extend the life of your IT Assets

Friday, April 16th, 2010

The economy has hit the IT departments of most companies, and budgets are under scrutiny. For organizations looking to save money, delaying the procurement of new assets can be a logical step. But it can be a tricky balance to squeeze as much life as possible out of existing assets without hazarding downtime. Here are some tips for boosting life expectancy without putting the enterprise at risk.
Focus On Power & Cooling
There’s always plenty of discussion about power and cooling issues in the IT industry, and for good reason.
Tweaking these areas can bring enormous boosts in energy efficiency and also prolong the life of IT assets.
Concentrating on power and cooling especially as high-density equipment is integrated into a center more often can keep IT assets not only rolling longer, but also working harder. One aspect of power and cooling that’s often overlooked is integrated coordination among seemingly minor data center items such as coordinate lighting, fire suppression, equipment placement, cable pathways, and air distribution, and that can create a bad design scenario. Focusing on how cooling and airflow affect the entire center can cause less strain on specific IT assets and prolong the life of that equipment as a result.
Take Advantage Of Virtualization
One approach to extending IT asset life expectancy will be virtualization. Along with that strategy, IT managers will have to focus on larger efficiencies, as well, such as implementing a more efficient storage strategy. High-density racks are here to stay and will become more of an issue for data center planners and operators over the next few years. While it will be important to remain effective at removing the associated heat, it’s also incumbent upon us as IT professionals to ensure that we are squeezing every bit of computing power and storage capability out of every watt. He adds that this means virtualizing workloads and weeding out the unnecessary over-storage of redundant information.
Put In The Extra Effort
One sacrifice that has to be made when extending IT asset life is the time it takes to do more planning,
monitor systems, talk with consultants and vendors, and ensure that assets are being maximized without risk. Time spent in monitoring is valuable, especially with sensor placement. Sensors can pick up changes in cooling, for example. What we see often is overcooling, in which managers are so concerned about overheating that they go to the extreme with cooling. That can be tough on the equipment, as well, just as much as overheating. In addition to monitoring, IT needs to put in the time to focus on lifecycle management, including the creation of end-of-life tracking for every asset. Although proper cooling, virtualization, and monitoring can extend an asset’s life, there does come a time when keeping an asset running past its expected end date can put a data center at risk. Setting an expiration date for each piece of equipment can be helpful for regular decommissioning rounds, which might include refurbishment and recycling, as well. Dedicating one staff member’s time to the effort can be useful, because that person can create a holistic view of asset management lifecycles and understand the complexities involved in monitoring and refurbishment. That person can also evaluate asset management software, which can pull together a complete view of assets, provide license tracking and compliance, and control IT budgets.
Establish Usage Measurements
Efficiently operating IT assets will last longer. In order to understand whether equipment is functioning
as efficiently as possible, it’s necessary to create a baseline, so that spikes in temperature or power
can be noticed sooner rather than later. Toward that end, establish a strategy that includes PUE (power usage effectiveness), a metric used to determine energy efficiency. It works by dividing the amount of power entering a data center by the power used to run the equipment inside the space. In other words, total facility power—which includes UPSes, battery backups, switch-gears, chillers, and CRACs (computer room air conditioners)—is examined in comparison to IT equipment power, covering servers, storage, and telco equipment. Sacco notes that in a properly designed space that’s running efficiently, the PUE should change very little from season to season, no matter how dramatic the temperature changes might be outside. An IT manager should also fold in measurement of individual pieces of equipment, such as servers.
Getting to this granular level can reduce consumption levels by examining compute cycles and optimizing
them where possible. Fewer compute cycles means that the server can operate for longer.
Extending asset life can be a complex endeavor, but if strong focus is put on strategies like
monitoring and virtualization, and awareness is created around efficiency and cooling, assets
can stay robust for longer life cycles.
Lock It Down
Working so diligently to extend the usability of an IT asset won’t do much good if that asset goes missing.
Consider the utilization of storage closets. Put in a couple different types of closets, especially ones that can be secured. You want to be able to store expensive things like laptops without worry, and you can also put extra servers in there, so you know they’ll be there if any problems come up and you have to swap them out.
Key Points

  • Focus on power and cooling efforts, because improper cooling can decrease the life of IT assets.
  • Look at SLAs to determine whether they’re meeting the needs of the data center.
  • Build in extra time for monitoring and planning to make lifecycle management a greater part of data center operations.