Uninterruptible Power Supplies
An Uninterruptible Power Supply can act as a first line of defence against data loss and downtime. They are affordable, easy to install and every practice should have at least one.
Australian medical IT systems are varied, comprising of a dizzying array of hardware and software combinations. However there is one device that should be compulsory in all medical centres regardless of software or hardware preference; the humble Uninterruptible Power Supply (UPS).
In its simplest form, a UPS is a battery that is plugged in between your mains power and your hardware (usually the server computer). In the event of a power failure, your server can continue running until the UPS battery is depleted. While this won’t allow your practice computers to run all day, it generally provides enough protection to ride out short power failures or safely shut down the computers.
UPS systems are designed to do one thing; provide a reliable, safe stream of power.
UPS devices not only help in the event of complete power failure, but also protect against brownouts and power surges. In fact it is these events that can be more damaging to the hardware than total power failure i.e. blackouts. In the absence of a UPS, a power failure will almost guarantee data corruption or loss. Severity may range from losing the current letter being transcribed, up to massive database corruption requiring a backup to be instated. In some cases, the corruption may not be detected at the time of the outage, which can lead to more severe problems down the track, especially for practices that don’t store permanent backups.
When purchasing a UPS, several questions need to be asked:
- How many devices need to be protected?
- How much power do these devices require to run?
- How long do these devices run for during a power outage?
In all cases, be conservative to allow for expansion and unrealistic manufacturer claims. With this information at hand, you can use one of the many web based UPS selection tools to choose an appropriate model (see “Quick Clicks”).
In a utopian setting, UPS devices would be deployed at each computer that either hosts or accesses data (generally all computers), however this is rare in practice. UPS devices are generally configured along side the practice software server, the justification being that this is where the important data the practice needs to protect resides.
Additional UPS devices may be deployed on client machines and network hardware to allow the entire system to function briefly during a power outage. As there shouldn’t be any critical data stored on the client machines, the UPS devices assigned to protect them won’t need to be as large or expensive as the one protecting your server.
In consultation with the product manual, even the most techno-illiterate practice staff member should be able to install and test a newly purchased UPS. After charging the UPS, all that needs to be done is:
- Shut the server down.
- Plug your server into the UPS.
- Start the server up.
Perhaps the only caveat to note is that many UPS devices have plugs for both battery backup and surge protection. These are not always clearly marked, so double check that you have used a battery-supported plug before starting your server.
More sophisticated UPS devices can be connected to the server via a data cable (USB, serial or network) and paired with software that can tell the server to safely shut itself down in the event of extended blackout. This is ideal as your server is protected even when your clinic is closed and the UPS runs out of juice.
Depending on your practice software, additional scripting may need to be performed to allow the UPS software to safely shut your database down. If the UPS software incorrectly closes the programs running on the server (i.e. by ‘force quitting’), the data corruption we are trying to avoid may eventuate. Your IT person in consultation with your practice software vendor should be able to assist if required.
While the benefits of UPS systems far outweigh any downsides, the following points should be given thought before purchasing:
If background noise in your consultation room is unacceptable, avoid the larger UPS devices that have audible cooling fans.
Keep the UPS (and server) in a well ventilated area and don’t inadvertently block the exhaust fans. Burning your surgery down may have adverse effects on the integrity of your data!
Protecting multiple devices with one UPS is quite possible, but the battery life will diminish in proportion to the extra power being drawn. Configuring automatic shutdown of multiple servers is also more complex.
Due to the large amounts of power used during laser printing, you should never attach a laser printer to a UPS. If you are averse to posting invoices and receipts, having a basic bubble jet printer available to connect to the UPS is the best option to get you through a blackout.
Maintenance and Batteries
Like all batteries, UPS cells deteriorate over time and should be checked at least once a month. While most devices have self testing functionality, nothing beats ‘flicking the switch’ (after a certified backup of your clinical data has been taken).
More robust products allow old batteries to be replaced, saving you the cost of having to purchase a new unit, and the hassle of learning to configure and install new software. The more expensive models even allow this change to occur with the server and UPS running, a feature that will be necessary for large 24 hour medical centres.
Don’t be tempted to stockpile batteries as they have a limited shelf life and may not perform well or at all when it comes time to install them. Purchasing a brand name UPS should ensure after sales support and parts are available when needed.
Posted in Australian eHealth