swh:1:snp:5115096b921df712aeb2a08114fede57fb3331fb
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
History
Tip revision: 19076c95aa2bcee55c26fcf0960cc844ad86ee9c authored by Levi Tamasi on 21 January 2021, 22:18:25 UTC
Update HISTORY.md for PR 7888 (#7890)
Tip revision: 19076c9
File Mode Size
db
doc
hdfs
helpers
include
port
scribe
snappy
table
thrift
tools
util
.arcconfig -rw-r--r-- 110 bytes
.gitignore -rw-r--r-- 61 bytes
AUTHORS -rw-r--r-- 193 bytes
LICENSE -rw-r--r-- 1.4 KB
Makefile -rw-r--r-- 8.7 KB
NEWS -rw-r--r-- 509 bytes
README -rw-r--r-- 1.7 KB
README.fb -rw-r--r-- 281 bytes
TODO -rw-r--r-- 494 bytes
build_detect_platform -rwxr-xr-x 7.6 KB
build_detect_version -rwxr-xr-x 1.1 KB
fbcode.gcc471.sh -rw-r--r-- 2.2 KB
fbcode.sh -rw-r--r-- 2.1 KB

README

back to top