[Beowulf] cluster softwares supporting parallel CFD computing
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.
Daniel Kidger daniel.kidger at clearspeed.comWed Sep 6 08:34:45 PDT 2006
- Previous message: [Beowulf] cluster softwares supporting parallel CFD computing
- Next message: [Beowulf] cluster softwares supporting parallel CFD computing
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Bogdan, Parallel applications with lots of MPI traffic should run fine on a cluster with large jiffies - just as long as the interconnect you use doesn't need to take any interrupts. (Interrupts add hugely to the latency figure) Daniel -----Original Message----- From: beowulf-bounces at beowulf.org [mailto:beowulf-bounces at beowulf.org] On Behalf Of Bogdan Costescu Sent: 06 September 2006 14:56 To: Mark Hahn Cc: Beowulf List Subject: Re: [Beowulf] cluster softwares supporting parallel CFD computing On Mon, 4 Sep 2006, Mark Hahn wrote: > 100 Hz scheduler ticks might make sense Please don't mention this to beginners (which is the impression that the original message left me) - if they care enough to search the information available on the subject they will either make their own mind or come back here to ask a specific question... Higher scheduler frequency was demanded, among others, by network developers who see the backlog run at this frequency - when this is too low, the network latency in certain situations increases which is undesired for most cluster applications. IMHO, low scheduler frequency is probably only good for mostly computational jobs (f.e. Monte Carlo style) where such latency doesn't matter and the small amount of time saved due to the lower number of context switches is used for real computations. -- Bogdan Costescu 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 _______________________________________________ Beowulf mailing list, Beowulf at beowulf.org To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
- Previous message: [Beowulf] cluster softwares supporting parallel CFD computing
- Next message: [Beowulf] cluster softwares supporting parallel CFD computing
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Beowulf mailing list
