First off, let me say I greatly appreciate your contributions to this board. It's people like you that can lead the group. I respect your perspective, but I am going to have to disagree.
It seems your main problem with servers is the complexity involved. Server hardware is pretty much invisible to the Operating System. It doesn't matter if you're using Windows 7, Server 2011, Windows XP, Windows 98. You have to take into account that the server is replacing all of the paper charts in your office when determining how much you want to spend.
People that complain about the price of the hardware. The way I figure it, it's way more expensive than the staff hours put into filing, stapling, hole punching, and all that paper and ink that goes into charts. Not to mention storing charts, getting shelves/boxes, and the space you lose to storing those charts (storage space). I'm pretty sure it's way way way less than the price of a legitimate server and switch setup. It's so much more efficient when retrieving and looking up information as well. I think a lot of people just lose sight of that and end up putting their entire practice on some $300-400 dollar desktop from the local best buy. Come on you probably spend more on ink for the month yet you're willing to wager your entire practice on some desktop from your local electronics store.
Think of server hardware of being like the really nice desktop hardware. At the core, that's all it is. The only complexity is when it's initially set up when you buy it so that argument doesn't carry much weight. What's going to happen?
- A power supply fails, but you have two and bright red light comes on that and the office keeps going. Replacing it is as simple as sliding out the bad unit and putting in a new one at your convenience. Your desktop power supply fails, you have to open up your entire computer and rewire it from scratch when you install a new one. And if you have no idea of how to diagnose a failure or installing a power supply, you're going to have to call emergency IT technician and you will spend much much more than the price of a redundant power supply.
- A hard drive fails and there's another one that's simultaneously writing the same exact data. You simply swap it and that's it. Your desktop hard drive fails and you're down until you can copy the data off your back ups onto the new computer. Install SQL, restore the databases, etc. If you have a lot of imported items, it can take even longer. All you have to do on a RAID array is plug in a new hard drive and press a rebuild button.
- You have a memory error and you get a blue screen and have to reboot. If you have ECC RAM, it corrects the error and the server stays on. Again, no user input involved here. It just works automatically..
- Server motherboards are designed to handle continuous stress. The quality control is better and some even test every board before they send them out to retailers/distributors. Again, no user input involved. I've seen plenty of desktop and especially notebook motherboards go bad. I rarely see server motherboards go bad. (Actually, I don't think I have. I even have a 10 year old PowerEdge server that's still going, running 24/7)
This is why I find that arguing about the complexity to be a bit overstated since you're essentially doing nothing, but you're getting so much out of it. You check some boxes on Dell's customize page. My recommendations can either be used to build your own or just match the boxes to the recommendation. It's even arguably simpler than having a plain desktop hardware if anything goes wrong as I demonstrated above. Servers aren't these magically super complex machines but just desktops with better parts.
.png "ACUF Donor I")
