In the event of a power issue, the right equipment saves the day. And your bottom line.
Sometimes it takes a glitch to highlight the importance of having the right equipment for the job. This month, we discuss some “must haves” that save businesses from expensive equipment damage, downtime and data loss.
A recent overnight Central Coast power outage affected over 145,000 PG&E customers, including more than 100 TekTegrity clients. Of more than 160 servers affected all but four were operating normally by 9am. We were able to get three of those online early in the day, some with minor repairs remaining, and one server literally bit the dust. That machine had already been flagged as an end-of-life-cycle machine, with replacement recommended. We were able to get a loaner server in by the end of the day.
This power outage was preceded by a brownout. The reduced voltage during a brownout can be especially damaging to electronics. Servers, designed to remain on at all times, are especially sensitive. Sudden shutdowns or low power situations can cause data corruption and hardware damage.
To prevent disaster and minimize interruptions, businesses should be prepared for this and other common situations.
External power failures
When the power goes out, those without critical 24/7 needs rely on an uninterruptible power supply, more commonly referred to as a “UPS.” A UPS provides alternate power for a short time, often around 10 minutes. This allows users to safely shut down their electronics. A UPS will not provide power during a long-term interruption.
Organizations with critical 24/7 needs also use UPS devices, but they are paired with alternate long-term power system solutions, such as permanently installed and regularly maintained generators and transfer switches. The UPS maintains power levels during the brief transition from utility company provided power to generator power.
Most of TekTegrity’s clients are setup with network managed UPS systems. A network managed UPS is different from a simple UPS that you can get at an office supply store. These more powerful UPSes protect from brownouts. They also allow a connected server to be shut down safely before the UPS battery drains completely. And after power comes back on, this type of UPS can wait till power stabilizes, then turn the server back on automatically.
Having smart UPS technology like this in place with many of our clients was what allowed us to get them back on line so quickly after such a widespread outage. Network managed UPS systems minimize downtime and avoid time-consuming travel for service calls. It’s easy to see how this saves money for businesses.
Any business reviewing recommendations for a new server should pay particular attention to the UPS line item. Prices can vary by several hundred dollars due to the type of UPS recommended. While network managed UPS systems cost more up front and require additional setup and configuration, we’ve never had anyone regret the investment after experiencing a power outage.
Internal power failures
Sometimes a power failure comes from within a server. A power supply is one of the most common computer components to fail. Yet, because they are specific to certain machines and manufacturers, and each business has a server selected to meet its own needs, most service providers are not able to stock server power supplies on their shelf. To get a replacement shipped out usually takes 24 hours.
A redundant power supply is a secondary power supply within a machine, capable of providing uninterrupted power if the primary supply fails. Not only does this protect the machine and its data, but the units are often “hot swappable,” meaning the failed component can be replaced without having to turn off the server or interrupt use. This is another great example of where a small up-front investment can prevent loss of data and save a day’s worth of productivity.
We would rather help our clients manage an occasional challenge than recover from an expensive business-crippling outage. And historically, that’s why every server upgrade we recommend includes a network managed UPS and redundant power supplies. These options can make a server a few hundred dollars more expensive than an entry-level system, but the cost is insignificant in the light of how much downtime can be reduced.
Russ Levanway is the CEO of TekTegrity, an IT Managed Services Provider serving the Central Coast and Central Valley. The organization’s Total Systems Management™ (TSM) service model provides preventative IT support at fixed monthly fee levels. For more information, visit www.tektegrity.com.