Servers are the heart of most modern SMBs. They host intranetwork data, customer data, and facilitate product delivery. Like any piece of equipment in your workflow, servers need to be carefully cared for in order to keep you up and running. With the strain that most businesses put on their servers, one of the most important maintenance variables is temperature management. Understanding why keeping your servers cool is vitally important and could save you from an expensive crash, troubling data loss or reduced hardware reliability.
How does temperature affect my servers?
Extreme temperature in server hardware can result in different forms of damage. Most SMBs see total failure as the most concerning outcome. A server that completely crashes for any reason results in costly data loss and service interruptions, but the unbiased advisory organization Uptime Institute warns about overheating that doesn’t result in total failure. Every 18 degrees higher than 70 degrees Fahrenheit, hardware reliability decreases by 50%. This decrease in reliability can be just as, if not more, expensive for your hardware budget in the long run.
Cooling methods can’t just be implemented and forgotten, they must be closely monitored to ensure the health of your server hardware–short and long term. Options for temperature management range from simple low-budget solutions to expensive outsourced alternatives; determining your server management budget will greatly depend on what types of methods you intend to implement at your SMB.
Which system you use to cool your server largely depends on how much power your hardware is using. The higher the watt, the harder it’s working. It will be easier to determine the scope of your temperature management needs when you have a thorough understanding of your power consumption.
PCWorld advises that simple conduction management is adequate for any equipment operating at less than 400 watts. This means simple solutions like positioning your server away from walls, low ceilings, cable clusters and anything else that can block hot air from dissipating naturally.
For watts between 400 and 2,000, strategic ventilation becomes a necessity. Adding passive ventilation is viable up to 700 watts, but fan assisted ventilation will be required above that up to 2,000 watts. With the increased power consumption, temperatures will rise and air movement needs to be more closely managed. At this stage simple vent and oscillating fans will suffice.
Anything higher than 2,000 watts needs to utilize dedicated cooling solutions. This means air-cooled units to actively reduce server room temperature. Depending on the size and arrangement of the space, a simple self-contained unit may be enough to reduce rising temperatures back into acceptable ranges. However, if you’re not sure, you should schedule a consultation with a vendor to consider more drastic cooling and monitoring methods.
Keeping your servers running at ideal temperatures means smoother data operations, lower hardware budgets and one less thing to worry about at your SMB. As your business continues to grow and develop, keep close tabs on increasing server loads–it could save you from devastating data loss. If you’d like more detailed advice about server management, or have any other questions about your hardware setup, contact us today.
Published with permission from TechAdvisory.org SOURCE