However, consider the global supplier situation for Opterons. Alas, Dell don't do them. Who would be the right supplier for this?
HP makes em and supports them (they even support Windows on them in an experimental beta, but I didn't tell you that). Additionally, Sun makes them, and they sell them VERY cheaply. That might be the best route. You don't gotta run Sun Linux 5.0 (or whatever they're calling it these days). Sun sold us (AOL) a dual opteron 242 for $2500 with 2gb of ram and 72gb of 15krpm disk not terribly long ago. Wikimedia doesn't exactly have AOL's relationship with Sun, but it might be possible to call and "talk them down" a little. And the FOSS argument doesn't apply to hardware, right? (because I know we have some Sun and Solaris haters here)
Oh, and don't forget the Gbytes and Gbytes of RAM!
For database servers? Yeah. 4gb minimum. If MySQL supports being built 64-bit, and is designed to take advantage of it, we can reap SUBSTANTIAL rewards from this. I'd say go with 8gb if we can afford it.
DBA's, speak up...
Am I missing something? Can we not just sync with {tick,tock}.usno.navy.mil? If usno.navy.mil goes down, we have much bigger problems than a toasted wikipedia.
Consider connectivity going down at a remote site (takes just one backhoe at the site 1/4 mile down the road where all of your supposedly "diverse" fibers join the same duct, or simply a router going mad, or someone hitting the Big Red Power-Off Button at ********* ***** [substitute your national Achilles Heel IXP]), rather than Global Thermonuclear War. Trust me, a local clock is a good thing. That's why it's such a useful service to have onsite.
I don't buy it. I've been syncing to USNO for years, and it Just Doesn't Happen to them. It doesn't happen to the DoD because DISA doesn't let them. NRO and {{spooky agencies}} all use them.
Now, that's a very good point. However, it might be better to do what Linus does and "let millions of people mirror it everywhere". This might be something to ask organizations like to UK Mirror Service, Internet Archive and Google to do on a formal basis. This way, the backups are
On a formal, hourly basis? Or do we decide we're comfortable with a day's loss of data if a nuke falls on florida?
off-site too. The current archive is 50G. If we take a week to back it up, that's a data rate of 50e9*8/(86400*7) = 662 kbps < 1 Mbps. So Wikipedia could perform complete backups weekly to three different sites at a cost of 3 Mbps sustained. That's cheaper than the cost of the tape media.
Weekly doesn't cut it. For full backups, it does. For incrementals, we need nightly or hourly. Does MySQL support live backups? (I understand on-the-fly transactions)
aa (talking systems shit and not getting flamed, yay)