Archives


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

[Beowulf] VIA C3 benchmarks?

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

Eugen Leitl eugen at leitl.org
Wed Jun 23 08:36:48 PDT 2004


On Wed, Jun 23, 2004 at 08:09:51AM -0700, Jim Lux wrote:

> I have just such a toy cluster.. it's 4 of the 533 MHz fanless units,
> however, we have only just started doing benchmarks on it.  It's running the

I'm rather impressed by the nano-ITX motherboard form factor -- it seems to
fit blade-like into 3U rackmounts, and they should be easy enough to cool. 
They also have compact flash sockets. On the minus side, the onboard NIC 
only does FastEthernet. Another plus:
http://www.linuxbios.org/status/index.html says via/epia and via/epia-m is
supported.

They seem largely suitable for web hosting (hardware acceleration for crypto,
hardware RNG, NX), but then http://linuxdevices.com/articles/AT2656883479.html says

"Q22: Beyond security, are other cool features planned?

A22: The next chip has some tools to do computationally intensive things
where hardware provides a big advantage. But I don't want to say yet what
they are.

Q23: Would they be useful for multimedia?

Yes, for multimedia, and for other things.

Q24: Like a DSP?

A24: Kind of like that."

which *does* sound interesting. 

> freevix configuration (since it supports the sound card hardware and is
> adapted to diskless booting) sans the x-windows stuff, but this uses the
> uclibc libraries and busybox for the shell stuff. I don't recall exactly
> which kernel version it is.

I'm pretty impressed by the busyboxy (I run Sveasoft for WRT54G which is
based on a MIPSel).
 
> The clustermatic folks also have a small cluster of these boards (they had
> them at the ClusterWorldExpo)
> 
> Suggest some standard tests and I'll run them in the next couple weeks,
> assuming I can build the benchmark successfully (this is by no means
> assured... the nodes have a pretty limited library set, and we're still
> thrashing through our tool chain issues)

-- 
Eugen* Leitl <a href="http://leitl.org">leitl</a>
______________________________________________________________
ICBM: 48.07078, 11.61144            http://www.leitl.org
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE
http://moleculardevices.org         http://nanomachines.net
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
Url : http://www.scyld.com/pipermail/beowulf/attachments/20040623/65154a19/attachment.bin


More information about the Beowulf mailing list