[Beowulf] SuSE 9.3
Lombard, David N
david.n.lombard at intel.com
Wed Jul 13 08:37:33 PDT 2005
From: Mikhail Kuzminsky Wednesday, July 13, 2005 8:18 AM
> In message from "Lombard, David N" <david.n.lombard at intel.com> (Wed,
> 13 Jul 2005 07:23:57 -0700):
> >From: Mark Hahn on Tuesday, July 12, 2005 5:27 PM
> >> > Corporate users and ISVs don't want to see the OS revised more
> >> > a year.
> >> which is sad, really. they've been so traumatized by the dominant
> >> platform that they expect that changing anything will break
> >> the very concept of a standard, let alone an interface standard
> >> is foreign to this mentality.
> >Theory v. practice. Implementation (i.e., compiler and linker
> >can substantially impact performance and correctness.
> >As example, the 2.4.6-2.4.18 range of kernels saw a steady rise in
> >short/small I/O performance at the expense of a steady and
> >loss of large-I/O performance.
> Sorry, does it incorrect for latest 2.4.x, for example 2.4.21 ?
Sorry, I didn't explicitly state this, but at 2.4.18 (maybe [.17,.19]?)
the large-I/O slowdown was corrected.
> BTW, if you say about ext2fs/ext3fs, and need huge I/O, why not to use
> xfs file system ?
Be assured, XFS *was* used, with very careful tuning to the app.
My comments represent my opinions, not those of Intel Corporation.
More information about the Beowulf