Mark Callaghan 33323f2111 Remove use of mmap for random reads
Summary:
Reads via mmap on concurrent workloads are much slower than pread.
For example on a 24-core server with storage that can do 100k IOPS or more
I can get no more than 10k IOPS with mmap reads and 32+ threads.

Test Plan: db_bench benchmarks

Reviewers: dhruba, heyongqiang

Reviewed By: heyongqiang

Differential Revision: https://reviews.facebook.net/D5433
2012-09-14 16:43:50 -07:00
..
2011-10-31 17:22:06 +00:00
2011-10-31 17:22:06 +00:00
2011-10-31 17:22:06 +00:00
2012-04-17 08:36:46 -07:00
2011-10-31 17:22:06 +00:00
2012-08-28 11:43:33 -07:00
2011-10-31 17:22:06 +00:00
2011-10-31 17:22:06 +00:00
2012-08-28 11:43:33 -07:00
2011-10-31 17:22:06 +00:00
2012-05-15 10:10:01 -07:00
2011-10-31 17:22:06 +00:00
2012-04-17 08:36:46 -07:00
2011-10-31 17:22:06 +00:00
2011-10-31 17:22:06 +00:00
2011-10-31 17:22:06 +00:00
2011-10-31 17:22:06 +00:00
2011-10-31 17:22:06 +00:00
2011-10-31 17:22:06 +00:00
2012-09-06 17:52:08 -07:00
2011-10-31 17:22:06 +00:00
2011-10-31 17:22:06 +00:00
2011-10-31 17:22:06 +00:00
2011-10-31 17:22:06 +00:00
2011-10-31 17:22:06 +00:00
2011-10-31 17:22:06 +00:00
2011-10-31 17:22:06 +00:00