Recently I was talking to some health care people about what it is that I do, and tried to put it into context that they would understand. The more I started to weave the fabric (pun intended) of what FCoE is and what it could mean in health care terms. The more I spoke the more their heads nodded and agreed. It got me thinking: could hospitals and healthcare really be the sweet spot for FCoE adoption?
Personally, I’m not in healthcare, have never worked in healthcare, and I confess I’ve never sold into the vertical. I do, however, have several friends and family who work within healthcare and by proxy I have come to hear the frustrations that they experience with healthcare IT.
From what I’m led to understand, most hospitals’ data centers are a hodgepodge of barely-working equipment that most modern data centers would have repurposed years ago. There are some notable exceptions (I’m told that Kaiser Permanente has an admirable IT model that they are looking to franchise out nationwide, e.g.) but they are in the extreme minority.
One of the reasons why EMR (Electronic Medical Records) systems and HIPAA have been so difficult to implement (and enforce) has been because the underlying infrastructure of hospital data centers have been the IT equivalent of Rube Goldberg machines.
Many EMR systems are DOS-based, specifically because there is no underlying coherence within the DC. DOS appears to be the only lingua-franca that is ubiquitous enough to ensure widespread implementation and compatibility. Indeed, the transfer of data from doctors’ and nurses’ stations more often than not looks like something out of Caractacus Pott’s workshop.
The reasons for this are pretty easy to understand. When you have a hodgepodge approach to acquiring equipment and force-fitting those square pegs into round holes, the mishmosh of hardware, software, networks, and storage, as well as the lack of resources, means that systems are put in place that have long outlived their usefulness (or ability to evolve into a more useful system).
In other words, they’re stuck.
But what if we were to take that liability and turn it into a future-proof asset? What if we were to put in the Data Center Esperanto that FCoE can be, and simplify the system? Pull the loose shoelaces tight?
Look, FCoE is a bolt-on technology. It’s designed to allow data centers to implement the technology without needing to Rip and Replace. But what if we turned that metaphor on its head?
FCoE can insert an single, specific infrastructure where one doesn’t exist yet.
In other words, what if we created an FCoE backbone to which existing infrastructures can attach, much the same way that ARPANET consolidated the various networks by interconnecting via TCP/IP protocol in the 1980s? ILLINET, FIDONET, BITNET, etc. don’t exist any more because they were attached to, and eventually superseded by, what was to become the Internet. We could do the same thing for hospitals and health care data centers.
Creating an FCoE backbone in a hospital would allow the existing LAN/SAN/NAS infrastructure – whatever it might look like – virtualize and converge the networks which would permit the deprecation of over-extended equipment, while finally permitting growth and evolution up to the application layer.
To me, the part of the FCoE adoption curve where the true breakthrough happens is when the new technology is used in ways that the designers didn’t necessarily intend. If we stick to the model of FCoE-as-Newcomer we may miss some of the truly useful cases where FCoE (and CEE by extension) can really shine.
It seems to me that hospitals and health care data centers, those mixed up, tangled up places where coherence is a pipe dream, might just be a sweet spot for FCoE.
You can subscribe to this blog to get notifications of future articles in the column on the right. You can also follow me on Twitter: @jmichelmetz