FW: [Beowulf] file IO benchmark
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.
Joachim Worringen joachim at ccrl-nece.deMon Nov 28 02:04:37 PST 2005
- Previous message: FW: [Beowulf] file IO benchmark
- Next message: FW: [Beowulf] file IO benchmark
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Imran Khan wrote: > Joe, > > I would have said the same about panasas but I like terragrid as every thing > is pretty much standard so not to many high over heads, plus it offers the > following: > > Standard Linux filesystem, and tools, so no re-training Imran, as far as I understood, some modifications to the file system *are* required to make it match with the consistency semantics of the Terragrid iSCSI target. Therefore, the descriptions I read always state that you can run ext2 via Terragrid, and nothing else. Are other file systems available now? What kind of patches are required (these should be open source, shouldn't they?)? > TerraGrid does not use a Meta Data Controller, so scales linearly. This is a nice point of the architecture; however, it is not clear that the MDC does in fact become a bottleneck. Lustre people, i.e., claim that thas has not been a problem even for large clusters. > TerraGrid is the only CFS solution with a 24X7 resilient option! What do you mean by 24x7? If a fs has redundancy, it will always work 24x7, doesn't it? Marketing pitch? > Increased reliability by support for diskless cluster nodes Do you consider this a special feature of the fs? You can achieve the same using NFS! Joachim -- Joachim Worringen - NEC C&C research lab St.Augustin fon +49-2241-9252.20 - fax .99 - http://www.ccrl-nece.de
- Previous message: FW: [Beowulf] file IO benchmark
- Next message: FW: [Beowulf] file IO benchmark
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Beowulf mailing list
