This machine will be splitting DB2 flat files for a living. That's pretty
much all it will do. The datasets range in size from a couple hundred K to
850 MB in size. The data are all on the SCSI drive at present, and plans
call for additional drives on this SCSI chain later on. I am running a home
made C application which does the job of sed and awk in that it reads in the
dataset file and splits it into files according to information in the first
field of each record, along with an assortment of error checking and
logging. So to boil it down, this machine needs to be able to manage many
thousands of file writes and reads very quickly. It already competes evenly
with a dual processor RS6000, but the boss would like to see just a little
bit more before we make a formal pitch.
I realize I have to improve the disk situation, but can you recommend any
specific things to tweak in the kernel that might push this box just a tad
faster?
Alan
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/