Revision 33323f21117e5bb8cf72569302f109aabb387bb9 authored by Mark Callaghan on 14 September 2012, 19:35:02 UTC, committed by Mark Callaghan on 14 September 2012, 23:43:50 UTC
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
1 parent fa29f82
| File | Mode | Size |
|---|---|---|
| bin | ||
| folly | ||
| gen-cpp | ||
| if | ||
| lib | ||
| libs | ||
| test | ||
| README | -rw-r--r-- | 927 bytes |
| assoc.h | -rw-r--r-- | 20.9 KB |
| openhandles.h | -rw-r--r-- | 5.9 KB |
| server.cpp | -rw-r--r-- | 1.1 KB |
| server_options.h | -rw-r--r-- | 6.0 KB |
| server_utils.cpp | -rw-r--r-- | 15.5 KB |
Computing file changes ...
