anyway if I force direct I/O, for example using oflag=direct in dd, the write performance drop as low as 8MB/sec

with 1MB block size. And each write it's about 120ms latency.

but that's quite a small block size.  do you approach buffered performance
if you write significantly bigger blocks (8-32M)?  presumably you're already
striping across OSTs?
lustre-discuss mailing list

Reply via email to