Storage systems used for EHR applications and PACS images have to ensure that vital patient information does not disappear, as such an event would certainly worsen patient outcomes and increase legal exposure. Storage systems for the last 30 years have used redundant arrays of disks (RAID), but RAID no longer works in this post-HIPAA world.
The two most prevalent types of RAID in use today are RAID5, which allows a single drive in the RAID group to fail without any loss of patient information, and RAID6 which protects against two drives in the RAID group failing. Health IT usually opts for RAID6. Why protect against two drives failing? Is that overkill? No, actually it is inadequate and reckless. Why?
When a drive fails, the bad drive is removed from the RAID group and a new drive is introduced. Then a long-running RAID rebuild process must happen before all of the data on the group of drives is truly safe again. On drives larger than 2TBs it commonly takes over 24 hours to complete a RAID rebuild. During that time, the remaining drives are being hit hard to extract all the data off of all their sectors, perform RAID math on that data, then return it in a coherent stripe across the RAID group.
Should another drive fail before the RAID5 rebuild is complete, all the data on all of the drives is gone forever. In a RAID6 rebuild, should another drive fail, all the data is still safe. But if two drives fail, all of the patient information is gone forever. IT goes scrambling for backup tapes and it takes days to restore vital patient information, assuming the restore works properly.
Ever make popcorn? Nothing, nothing…pop, then pop-pop-pop-pop. Well, all the drives in a RAID group were probably made on the same manufacturing line, so when one of them is worn out, the peer drives are also approaching their end of life. If your data matters, it should be protected with advanced erasure coding.
In advanced erasure coding, M of N drives need to be alive for data to be fully protected. Should any number of drives fail, all of the nodes in the Swarm work in parallel to heal the situation as quickly as possible so data is fully protected once again.
The result? Advanced erasure coding recovers in a fraction of the time and provides more protection than RAID. The advanced erasure coding feature is integral to Caringo Swarm software-defined object storage. Swarm is fully compatible with the Health IT systems commonly in use today. Check out our case studies to see how we have helped St. Jude Children’s Research Hospital, Johns Hopkins University, and the Department of Defense.