FNN vs GigabitEther & Myrinet
bogdan.costescu at iwr.uni-heidelberg.de
Thu Oct 25 05:16:16 PDT 2001
On Wed, 24 Oct 2001, Greg Lindahl wrote:
> One big problem with both the GAMMA and MVIA projects is that
> originally, neither was doing any kind of error correction, simply
> telling the application "whoops! there's a problem." That doesn't work
> with MPI programs. Now both have added error correction. But they both
> still have the problem of the fast changing market in gigE chipsets,
> and neither one has any scheme for directly plugging into BSD or Linux
> kernel drivers.
Why don't they get some support from the big guys ? For example, I
always wondered after Scyld announced its distribution if they are going
to include some kind of low-latency package, be it GAMMA, MVIA or
something else... Scyld certainly has the knowledge (but I don't know
about resources) to make it happen.
Plugging in directly in the kernels is not so easy; low-latency is not
only about transmission and reception of packets, but also about memory
management, especially pinning down memory segments from the kernel side
in order to make them available to the user side and the other way around.
However, IMHO doing this in Linux at the moment is not future-proof; apart
from the ever changing VM subsystem, some people favour iovecs to do
kernel-userland data passing; OTOH when the zero-copy network changes were
introduced (around 2.4.3 or so), there were discussions about using iovecs
- some networking guys said that they are too complex and time-consuming
to set up and use and so they introduced yet another mechanism which is
now specific to networking...
IWR - Interdisziplinaeres Zentrum fuer Wissenschaftliches Rechnen
Universitaet Heidelberg, INF 368, D-69120 Heidelberg, GERMANY
Telephone: +49 6221 54 8869, Telefax: +49 6221 54 8868
E-mail: Bogdan.Costescu at IWR.Uni-Heidelberg.De
More information about the Beowulf