Hi,
given the following job file ...
--- snip ---
[global]
ioengine=sync
direct=0
# Block sizes for I/O units: 25% 19k, 15% 177k, 60% 350k
bssplit=19k/25:177k/15:350k/60
# Use mixed workload: 30% random IO, 70% sequential IO
percentage_random=30
size=${SIZE}
numjobs=${NUMJOBS}
runtime=${RUNTIME}
directory=${DIRECTORY}
[application]
# Define of I/O pattern: Random read.
rw=randread
--- snip ---
This is run with the following command:
$ RUNTIME=0 NUMJOBS=4 SIZE=4096m DIRECTORY=/work/testsoft fio jobfile.fio >fio.out 2>&1
I have noticed from the output file, that this results in different
amounts of data read by the individual processes:
$ grep io= fio.out
read : io=5847.5MB, bw=149458KB/s, iops=627, runt= 40063msec
read : io=4096.2MB, bw=140358KB/s, iops=595, runt= 29884msec
read : io=4096.3MB, bw=140889KB/s, iops=596, runt= 29772msec
read : io=5246.4MB, bw=134821KB/s, iops=560, runt= 39847msec
READ: io=19286MB, aggrb=492947KB/s, minb=134820KB/s, maxb=149458KB/s, mint=29772msec, maxt=40063msec
I have expected that every individual process will read
its 4096 MB and then stop further reading. Or am I missing
something?
Thanks in advance.
Best regards,
Juergen
--
To unsubscribe from this list: send the line "unsubscribe fio" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
