From: nmm1@cus.cam.ac.uk (Nick Maclaren)
Newsgroups: comp.parallel.pvm,comp.parallel.mpi
Subject: Re: Switching HUB for beowulf cluster?
Date: 4 Feb 1999 11:36:10 GMT
Organization: University of Cambridge, England
Message-Id: <79c0na$hk5$1@pegasus.csx.cam.ac.uk>
References: <78osd7$9ji$1@news.kren.nm.kr>
    <78piue$lrv$1@pegasus.csx.cam.ac.uk> <36B966E6.F5E27A60@dolphinics.com>
Originator: nmm1@taurus.cus.cam.ac.uk
Xref: ukc comp.parallel.pvm:7971 comp.parallel.mpi:4588


In article <36B966E6.F5E27A60@dolphinics.com>, James Cownie <jcownie@dolphinics.com> writes:
|> Nick Maclaren wrote:
|> 
|> > |> 2. How much is the need for 100Mbps bandwidth for a beowulf cluster?
|> > |> Will 10M bps only switching hub be ok for clustering or
|> > |> will 100M bps switching hub lead to better overall performance?
|> > 
|> > In terms of latency, there is no difference.  In terms of bandwidth,
|> > a lot.  As the extra expense of 100M bps is now small, I advise
|> > going for that.
|> 
|> The reason for using a switching hub, rather than a simple contention
|> hub
|> is to increase the bi-sectional bandwidth.
|> 
|> The latency difference (between teh two hubs) is likely unmeasurable at
|> the
|> MPI level, and the point to point bandwidth will also be the same. 
|> However a switching hub will allow many simultaneous communications.
|> This can be very important for the overall performance of your
|> application.

Yes, but he was asking about the difference between 10baseT and
100baseT switching hubs!

The current cost-effective solution is a 100baseT switching hub.
For better bandwidth (but not latency) at a higher cost, Myrinet,
HiPPI, Giganet etc. are useful.  For peak bandwidth, GSN (aka
HiPPI 6400) is the only game in town.


Regards,
Nick Maclaren,
University of Cambridge Computing Service,
New Museums Site, Pembroke Street, Cambridge CB2 3QG, England.
Email:  nmm1@cam.ac.uk
Tel.:  +44 1223 334761    Fax:  +44 1223 334679