test_fsync output improvement - Mailing list pgsql-hackers

From Bruce Momjian
Subject test_fsync output improvement
Date
Msg-id 201007040244.o642i9N07887@momjian.us
Whole thread Raw
List pgsql-hackers
I have modified src/tools/test_fsync so its output is operations per
second, instead of total seconds, e.g.:
Loops = 10000Simple write:        8k write                      52549.436/secondCompare file sync methods using one
write:       (open_datasync unavailable)        open_sync 8k write              575.599/second        (fdatasync
unavailable)       8k write, fsync                 349.669/secondCompare file sync methods using two writes:
(open_datasyncunavailable)        2 open_sync 8k writes           215.472/second        (fdatasync unavailable)
8kwrite, 8k write, fsync       566.493/secondCompare open_sync with different sizes:        open_sync 16k write
   247.207/second        2 open_sync 8k writes           219.824/secondTest if fsync on non-write file descriptor is
honored:(Ifthe times are similar, fsync() can sync data writtenon a different descriptor.)        8k write, fsync,
close         719.469/second        8k write, close, fsync         1457.288/second
 

This should make the values more useful.

--  Bruce Momjian  <bruce@momjian.us>        http://momjian.us EnterpriseDB
http://enterprisedb.com
 + None of us is going to be here forever. +


pgsql-hackers by date:

Previous
From: Robert Haas
Date:
Subject: Re: reassign owned to change the ownership for op class and family
Next
From: subham@cse.iitb.ac.in
Date:
Subject: Needs Suggestion