Before a thread (in May 2015), below strange output.

>dd if=/dev/zero of=/some/path/speed_test.dat bs=2M count=1
1+0 records in
1+0 records out
2097152 bytes (2.1 MB) copied, 33.6993 s, 62.2 kB/s


I try a check the suggestions.

Check same condtions.

>dd if=/dev/zero of=/mnt/data1 bs=2M count=1
1+0 records in
1+0 records out
2097152 bytes (2.1 MB) copied, 192.439 s, 10.9 kB/s

Increase cachesize.

>dd if=/dev/zero of=/mnt/data1 bs=2M count=1
1+0 records in
1+0 records out
2097152 bytes (2.1 MB) copied, 0.0736621 s, 28.5 MB/s

Decrease bs (cacheseize 1024k).

>dd if=/dev/zero of=/mnt/data1 bs=128k count=16
16+0 records in
16+0 records out
2097152 bytes (2.1 MB) copied, 123.385 s, 17.0 kB/s

In samll cachesize, s3ql requests a number of PUT/DELETE methods to backend.
This is not might it be slow cause?

-- 
You received this message because you are subscribed to the Google Groups 
"s3ql" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to