Re: [PATCH] concurrent inode allocation for ext2 against 2.5.64

[William Lee Irwin III <wli@holomorphy.com>]

On Sat, Mar 15, 2003 at 02:37:18PM -0800, Andrew Morton wrote:
>> Looks like it's gone nuts when 128 processes all try to close lots of
>> files at the same time.
>> One possible reason for this leaping out is that all the instances are now
>> achieving more uniform runtimes.   You can tell that by comparing the dbench
>> dots.

On Sat, Mar 15, 2003 at 02:58:42PM -0800, William Lee Irwin III wrote:
> For some reason this version of dbench doesn't produce dots. I logged
> what it did produce, though. It looks something like this:

There's a problem with the old dbench:

Throughput 82.0899 MB/sec (NB=102.612 MB/sec  820.899 MBit/sec)
$ (time ./dbench/dbench 128) |& tee -a ~/dbench.output.15
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++128 clients started
********************************************************************************************************************************
Throughput 1841.38 MB/sec (NB=2301.72 MB/sec  18413.8 MBit/sec)
./dbench/dbench 128  73.31s user 6.75s system 1802% cpu 4.440 total