[HCoop-Discuss] hardware advice

[Adam Megacz]

Adam Chlipala <adamc at hcoop.net> writes:
> it sounds like you're saying that disks are the only kind of
> hardware that we need to accept as having failure rates high enough
> that we need to contemplate dealing with failures at anything but a
> leisurely pace.  Am I interpreting you right?

No; I'm saying that RAID and hot-swap bays make disks the only place
where it's sensible to buy cheap-and-unreliable parts.

> Also, what makes two the magic number of machines to have available at a 
> time?

It's "minimum", not "magic". :) Right now we are well below both the
minimum number and (if such a thing exists) the magic number.

> Adam, which model numbers of IBM and Sun servers do you think are the 
> "sweet spot" choices to buy right now, and where did you get such a good 
> price on your beefy IBM server? :)

eBay.  IBM's model numbers are confusing, but if it has "eServer" (not
"xSeries") in the name and has a pair of opterons, it's close enough.

> Do you think that the KVM feature of the service processor that you 
> mentioned before is enough that we should prefer IBM over Sun overall?

Not by itself.  Also, Sun hardware tends to be fairly expensive, even
used.  If you're on a tight budget the older IBM stuff tends to be a
better choice.

> You can probably tell that I'm pretty ignorant about the nuts and
> bolts of hardware and hosting today.

While the pejorative "nuts and bolts" may be an accurate description
of these issues, they are what drives the co-op's top-level
architectural decisions.  Perhaps there are some lurking meta-level
issues in the way hcoop's governance structure allocates
responsibility (speaking of which, is ntk still breathing?) vs
authority.

  - a