Archives


- Beowulf
- Beowulf Announce
- Scyld-users
- Beowulf on Debian

[Beowulf] Re: Beowulf Digest, Vol 19, Issue 33

Many of your questions may have already been answered in earlier discussions or in the FAQ. The search results page will indicate current discussions as well as past list serves, articles, and papers.

Search

Maurice Hilarius maurice at harddata.com
Mon Sep 26 05:11:08 PDT 2005


Bruce Allen wrote:

>...
>Spending the $$$ to buy some extra nodes won't work in our case.  We don't 
>just use the cluster for computing, we also use it for data storage. 
>Each of the 400+ nodes will have four 250GB disks and a hardware RAID 
>controller (3ware 9500 or Areca 1110).
>
AMCC (3Ware) have started shipping their new 9550X series.
Big changes are included:
True PCI-X, separate channel chips for each drive, new RAID CPU.
You may also want to look at the offerings from  LSI.
300X supports S-ATA2, and their SAS RAID cars are just about ready to
ship to the general public.
Nice thing with that, of course, is downward compatibiloity with S-ATA1
and S-ATA2  as well.

>  If a node is acting odd, we'd like 
>to be able to diagnose/fix/reboot/restore it quickly if possible.  To 
>replicate the data from a distant tape-backed repository will take many 
>hours. So having some 'extra' machines doesn't help us so much, since we 
>wouldn't know what data to keep on them, and moving the data onto them 
>when needed would normally take much longer than bringing back to life the 
>node that's gone down.
>
>Cheers,
> 	Bruce
>  
>
IPMI cards should do the trick.
Most decent motherboard makers offer that now for around $50 - $100 per
node.

>
>On Sat, 24 Sep 2005, Douglas Eadline wrote:
>
>  
>
>>>We're getting ready to put together our next large Linux compute cluster.
>>>This time around, we'd like to be able to interact with the machines
>>>remotely.  By this I mean that if a machine is locked up, we'd like to be
>>>able to see what's on the console, power cycle it, mess with BIOS
>>>settings, and so on, WITHOUT having to drive to work, go into the cluster
>>>room, etc.
>>>
>>>      
>>>
Also, if it is just fan and temperature monitoring that is needed,
add-on boards to do that run at $80 per node, and can be easily
retrofitted..

Lastly, as most server/cluster boards have serial console option, it is
no pretty easy to add serial to ethernet converters to provide remote
I/O without the expense involved in some of the fancier terminal server
boxes.
At around $30 - $65 per port devices like this do the job quite well:
http://www.lavalink.com/index.php?id=263
http://www.lavalink.com/index.php?id=262



-- 
With our best regards,

Maurice W. Hilarius
Hard Data Ltd
Edmonton, AB, Canada       http://www.harddata.com/
 



More information about the Beowulf mailing list