<html><head></head><body>Thanks a lot. I'm looking forward to test this. Wounder why I haven't stumpled across this issue. While HPC workloads try to avoid syscall overhead, there are lots of software that rely on the OS to cope with small IOs. Think the average user will become a lot happier.<br>
<br>
Cheers,<br>
Hans Henrik<br><br><div class="gmail_quote">On October 2, 2016 11:21:55 PM GMT+02:00, "Dilger, Andreas" <andreas.dilger@intel.com> wrote:<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
<pre class="k9mail">Please test with Lustre master (pre-2.9.0). There were optimizations landed specifically to improve small file read performance. While Lustre 2.9.0 isn't released yet, it is getting very close. <br /><br />Cheers, Andreas<br /><br /><blockquote class="gmail_quote" style="margin: 0pt 0pt 1ex 0.8ex; border-left: 1px solid #729fcf; padding-left: 1ex;"> On Oct 2, 2016, at 06:54, Hans Henrik Happe <happe@nbi.ku.dk> wrote:<br /> <br /> Hi,<br /> <br /> While testing small sequential reads I noticed that Lustre is more than 10 times slower than local fs when reading cached data. So basically no network I/O to OSSes. I wanted to check if readahead was working for this small I/O case, but it seems that cached case isn't.<br /> <br /> The client is running Lustre 2.8 and CentOS 6.8.<br /> <br /> Lustre:<br /> <br /> $ dd if=file of=/dev/null bs=512<br /> 2097152+0 records in<br /> 2097152+0 records out<br /> 1073741824 bytes (1.1 GB) copied, 20.5081 s, 52.4 MB/s<br
/> <br /> $ dd if=file of=/dev/null bs=4k<br /> 262144+0 records in<br /> 262144+0 records out<br /> 1073741824 bytes (1.1 GB) copied, 2.91177 s, 369 MB/s<br /> <br /> $ dd if=file of=/dev/null bs=1M<br /> 1024+0 records in<br /> 1024+0 records out<br /> 1073741824 bytes (1.1 GB) copied, 0.160732 s, 6.7 GB/s<br /> <br /> Local fs:<br /> <br /> $ dd if=/tmp/file of=/dev/null bs=512<br /> 2097152+0 records in<br /> 2097152+0 records out<br /> 1073741824 bytes (1.1 GB) copied, 1.56432 s, 686 MB/s<br /> <br /> $ dd if=/tmp/file of=/dev/null bs=4k<br /> 262144+0 records in<br /> 262144+0 records out<br /> 1073741824 bytes (1.1 GB) copied, 0.275451 s, 3.9 GB/s<br /> <br /> $ dd if=/tmp/file of=/dev/null bs=1M<br /> 1024+0 records in<br /> 1024+0 records out<br /> 1073741824 bytes (1.1 GB) copied, 0.148798 s, 7.2 GB/s<br /> <br /> <br /> Is this a known issue?<br /> <br /> Cheers,<br /> Hans Henrik Happe<br /><hr /><br /> lustre-discuss mailing list<br /> lustre-discuss@lists.lustre.org<br
/> <a href="http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org">http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org</a><br /></blockquote></pre></blockquote></div><br>
-- <br>
Sent from my Android device with K-9 Mail. Please excuse my brevity.</body></html>