Health-care enterprises small and large continue to struggle with accommodating an ever-growing archive of imaging data—and efficient, long-term storage of these data is just part of the challenge. ImagingBiz.com spoke with Chuck Christian, CIO at Good Samaritan Hospital, Vincennes, Indiana, about the community hospital’s approach to storing and protecting its medical-image data.
ImagingBiz.com: How much image data did Good Samaritan Hospital generate last year?
Christian: We generated about 90,000 exams.
ImagingBiz.com: How much image data are you archiving, and what is the year-over-year growth rate since you acquired PACS?
Christian: We’re archiving everything—we save everything. We probably add one or two terabytes a year, and we add them in blocks. I’m hoping our growth will continue to increase, because that means our business is increasing. We recently signed an agreement with our PACS vendor to incorporate all of our cardiac-catheterization laboratory images into the archive, so our volumes are definitely going to grow.
The biggest challenge, for us, is making sure we have enough money to pay for everything: Archiving is one thing, but backing it up (and making sure you can restore it) is another. Creating the archive is just the tip of the iceberg. Improving how fast we can restore it starts to get costly. You try to find a good balance between cost-effective backup storage and being able to recover quickly. It keeps us on our toes. We’re currently looking at off-site storage and recovery to see what might be most appropriate and cost effective.
ImagingBiz.com: When did you begin considering a disaster recovery/business-continuity plan for Good Samaritan Hospital’s image data? Was it triggered by an event or by legal and regulatory concerns?
Christian: It’s just one of the things we looked at when we first bought the PACS. We knew anything we put online, we would need to back up in the event of a failure. It was prudent: We were making sure we did everything we were supposed to do. ImagingBiz.com: Describe your current solution. How do you define sound disaster recovery/business-continuity planning, and what would your ideal solution be?
Christian: We use an industry-standard backup solution for all of our applications. When we originally bought our PACS, it came with a backup solution, which was separate; we rocked on with that for a while, but eventually, we migrated it over to our current backup solution, which is tape based. I would like to get to a remote SAN backup, but it’s more than we can afford at this time, so we’re looking at more cost-effective solutions for that kind of replication.
The dream scenario would be having images automatically duplicated at the same time and stored somewhere else, so the secondary data center would be connected at high speed, but it would also have to be cost effective. As a community hospital, we’re in a place where we may have to partner.
We’ve considered doing some colocation with Vincennes University, which is 20 or 30 miles away, but is that far enough? We’ve looked at data centers in Indianapolis and Evansville and Columbus. The options are there; it’s just a matter of finding the right balance. It’s like insurance in that you need it, but you hope you never have to use it.
The other thing I’ve come to realize is that my job is to identify the risks for the organization and offer potential solutions, but it’s the administration’s responsibility either to accept or to mitigate the risks. I have to leave those decisions to them. Making calls about what the organization can and cannot do is not within my realm.
ImagingBiz.com: Has your solution changed since its inception? If so, how—and what did you learn that prompted those changes?
Christian: The biggest change was moving away from the solution that came with the PACS. We wanted to get it standardized. If we do have a disaster, we felt it would be far easier to have one solution to restore from, rather than multiple sources.
ImagingBiz.com: How quickly can your solution kick in, should there be an archive failure?
Christian: How quickly it would start depends. If there were a catastrophic failure in the servers, it could take a day to get parts in here. It just depends on a number of factors. It would take a while to restore everything; you can only spin from tape so fast. The other thing to remember is that just because you have a backup solution, that doesn’t mean you’re