getting data to nodes....
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.
Dean Johnson dtj at uberh4x0r.orgFri Feb 14 11:43:20 PST 2003
- Previous message: getting data to nodes....
- Next message: getting data to nodes....
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
On Fri, 2003-02-14 at 12:38, Mike Eggleston wrote: > On Fri, 14 Feb 2003, ambrose lewis wrote: > > > Hi All: > > I'm thinking about using an MPI code to analyze a > > large remote sensing dataset. > > I'm wondering the best way to distribute the data to > > the individual nodes (NFS, MPI message, other)???? > > Is there a "best practice" to handle this? > > I use pvm to send messages to my nodes to read a specific > datafile that resides in a NFS filesystem. > Depending on the size of the data, nature of the data, the network setup, and how your application uses the data, this can be a real bad thing. For instance, if it broadcasts the message "go grab your data", all the nodes will gang tackle the nfs server, to your detriment. If your application staggers the messages to get the data, then you would be *much* better off. -Dean
- Previous message: getting data to nodes....
- Next message: getting data to nodes....
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Beowulf mailing list
