Don’t roll the dice on data loss
Implement smart recovery to reduce disaster recovery costs in healthcare. By Chris Poelk er
ealthcare IT professionals know that data loss is simply not an option for them. Losing power or losing data can mean losing important docu- ments that affect doctors’ decisions. Despite
this fact, many healthcare IT fi rms don’t have a successful disaster recovery (DR) plan in place. More often, the DR plan hasn’t been tested in years.
The truth is that backup is a nuisance, and shrinking IT budgets are pushing DR plans to the back burner. But of all people, healthcare professionals should understand the value of an insurance policy. And an insurance policy is just like a DR plan. Cutting corners on your insurance policy or DR plan may seem like it will save money, but the cost will come back two- fold in the event of a disaster. That may leave many wondering, “How expensive is a disaster in the healthcare market?” According to publicly available data from a few years ago, a healthcare provider will lose $640,000 for every hour its systems are down. If data is an IT department’s lifeblood, an hour of data loss is like hitting the jugular.
Do you know how hard your IT department will be hit in the event of a disaster? From a hurricane to a simple power outage, this is a question you need to know. Take a look at these issues to help analyze your backup and DR costs and avoid that frightening six-fi gure dollar amount.
Managing tape with de-dupe With tape, your goal should be to use it as an archive stor-
age tier. This will allow you to minimize the need to recover from tape. Tape gets a bad rap mostly because it’s terrible for recovery. However, it’s great for archiving. Using a traditional backup operation of daily incremental and weekly complete backups, you need to manage around 100 terabytes (TB) of tape media for every 20 TB of production data with a two percent change and a three percent growth rate. And that’s only for a simple fi ve-week retention plan. Simple math shows that a shop with similar retention requirements and 100 TB of data would need 550 TB of tape (100 x 110 = 550). If you add de-dupe to the mix, you can store the same capacity in only 28 TB of disk at 20:1, and just cut a tape copy once a month.
14 March 2012 Calculating infrastructure and time
Don’t forget that the performance metrics need to include the entire data path: network, backup server, storage and backup target (tape or disk) as a whole. It doesn’t matter how fast your backup target is if you can’t feed it fast enough. Re- member that the time to recover using traditional approaches is usually twice the time to back up. As an example, a small virtual tape solution could provide hundreds of virtual LTO4 drives over Fibre Channel for a fraction of the cost.
Time needed to back up your data center
The current capacity of all the applications needing protection
The performance in terabytes per hour of your current backup solution
Managing your network Your wide area network (WAN) needs to be able to handle the data change happening on a day-to-day basis. It should be measured for peak loads so that critical batch-processing periods, such as end-of-month or end-of-year processing data,
HEALTH MANAGEMENT TECHNOLOGY www.healthmgttech.com