It's all too likely that this news will go completely unreported in the mainstream news as they would rather report on the latest new network TV show launches and lunatics (excuse the pun) rambling about civil war over healthcare reform.
More on the discovery here.
]]>Two Identical HP DL580 G5 Servers
It would be reasonable to ask what these are being used for. They are VMware Cluster Nodes. I expect to be able to consolidate along the lines of 50-70 VMs accross both nodes from prior experience with as many as 100 (though I'm not a fan of that level of vertical build on VM nodes in case of hardware failure).
This is one of the reasons I love supporting VMware and virtualization host systems in general. Because of the licensing (per processor socket rather than per core) for VMware the most cost effective hardware is often on the upper end of processor speed and memory capacity limits resulting in truly monster boxes.
]]>The first of which is almost a total lack of knowledge about HIPPA, and how it should affect information security with hospitals. Its not that IT staff don't want to comply but rather that they do not have the training or the information available to make a good attempt at it. Where is the well defined external audit process to encourage upper management to spend the money on information security? If you haven't put much thought into it yet, that is the real reason you should enforce multiple yearly audits on healthcare IT. Management would rather spend the money on the revenue side for new medical equipment and staff than on securing the core IT infrastructure.
I too often see a "slap a bandage on it" approach to how systems and network security is handled in these environments. The only time it comes to the forefront is when an organization makes front page for losing patient data, and forgotten by everyone including the public a couple weeks later. The public is bombarded with news stories on information security breaches from hundreds of companies a year, many of which they've done business with. My belief is that consumers (and me for that matter) just go under the assumption that its part of life and an acceptable risk and rarely bother to take their business elsewhere unless they personally get burned (and know which company caused it).
I can't speak on a national level though I've heard enough information second-hand to believe that this is more of a systemic issue within healthcare rather than the exception. It makes sense from a growth standpoint as these organizations were still on mostly paper record systems up until only a couple of years ago. Even if they had a medical records system they were never fully integrated with every aspect of patient care like they are now. Organizations are still dealing with the rapid IT growth pains caused by massive healthcare demand increase coupled with a need to stretch medical staff further by increasing the number of patients they can see in one day. This rapid growth has resulted in almost complete dependance on IT systems by medical staff. What I mean is not that a Doctor is incapable of doing diagnostics on a patient but rather this new batch of residents do not know how to fall back on paper forms and manual procedures for administrative tasks that used to be common knowledge. In many cases incoming staff are given a quick train-in on the manual procedures and then sent off to a week of training on using the medical record system. It is ludicris to expect that in a true IT disaster where the EHR was down that these same staff members wouldn't be massively hampered by having to figure out the manual process as they go along?
While looking around I noticed another good blog covering this topic:
http://hcrenewal.blogspot.com/2009/06/hospital-computing-backwater-and-need.html
]]>I plan to run vSphere on it and will be pointing it to my Solaris file server via iSCSI. I plan to add more memory as needed down the road. The server will take up to 32GB and will have 4 slots free the way I ordered it. Pics to come soon.
Quick specs on my fileserver:
I have rolling rotating snapshots on for the entire RAIDZ storage pool with snaps every 15 minutes retained for a day, daily snaps retained for a week, weekly snaps retained for a month and monthly snaps retained for a year. It has saved me the loss of files I've accidentally deleted or modified several times. Of course its not a backup strategy without offsite copies which is where I use Amazon S3 storage for my most critical files.
With the performance I've seen out of iSCSI on an untuned Windows box pointed to this file server I'm confident that I could easily scale it into a larger case with more drives as needed and for much cheaper than an actual single use iSCSI storage appliance/array. It's definately opened my eyes to the compabilities of software RAID when properly implemented on decent (but cheap) hardware.
]]>