On Mon, 10 Apr 2006 at 3:10pm, David Thompson wrote
> Joshua Baker-LePain wrote:
> Yeah, what's up with SuperMicro and Opteron systems? Looking at their web
> site, it's like they have disavowed any knowledge of having built anything
> Opteron. Did they back out of Opteron, or do you have to be a member of a
> secret society to get to their information?
No, you just need to know the code word -- Aplus. As in
<http://www.supermicro.com/Aplus/>.
>> One caveat with these Supermicro boards is the the onboard SATA (some
>> sort of Adapted branded Marvell thingy) isn't supported. They offer some
>> sort of binary download, but we all know how much fun that is. I just put
>> 2 port 3wares in 'em all (they're cluster compute nodes). Sure, it's
>> expensive for a simple SATA controller. But it's also no fuss and rock
>> solid. The newer SM boards have HT2000/HT1000 based SATA which centos 4.3
>> should support, but I've no experience with 'em.
>
> Yeah, I've been there also. We also use alot of threeware in these
> applications. The other add-in sata that I've used in 3112, but that hasn't
> always worked.
Exactly. If it doesn't always work, then it's not worth spending time on.
>> I don't know how deep my knowledge is, but we run a lot of memory
>> intensive code on these nodes and get a 2X speedup with dual CPUs. So,
>> IME, there's not much penalty with the extra hops to get to the other
>> CPU's memory.
>
> Yeah, it's cheaper to buy two CPUs and use 1GB dimms (8 slots) compared to 1
> CPU and 2GB dimms (4 slots). But the CPUs aren't going to be loaded (at all),
> so I'm concerned of hurting throughput by splitting the memory space. Tyan is
> working on a mobo with 8 slots per CPU (S3892), but my favorite vendors aren't
> carrying it (yet).
I strongly doubt you'd be hurting throughput with 2 CPUs vs. 1.
--
Joshua Baker-LePain
Department of Biomedical Engineering
Duke University
