Bug 682831

Summary:	Bad ext4 sync performance on 16 TB GPT partition
Product:	Red Hat Enterprise Linux 6	Reporter:	Debbie Johnson <dejohnso>
Component:	kernel	Assignee:	Eric Sandeen <esandeen>
Status:	CLOSED ERRATA	QA Contact:	Eryu Guan <eguan>
Severity:	high	Docs Contact:
Priority:	medium
Version:	6.0	CC:	anton, eguan, esandeen, kkolakow, knweiss, kzhang, mishu, ndoane, pasteur, rmusil, roland.friedwagner, rwheeler, yugzhang
Target Milestone:	rc	Keywords:	Regression
Target Release:	---
Hardware:	x86_64
OS:	Linux
Whiteboard:
Fixed In Version:	kernel-2.6.32-128.el6	Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:	572930
Clones:	1020728 (view as bug list)		Environment:
Last Closed:	2011-05-23 20:43:06 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:
Bug Depends On:	572930
Bug Blocks:	1020728

Comment 2 RHEL Program Management 2011-03-07 18:40:36 UTC

This request was evaluated by Red Hat Product Management for inclusion
in a Red Hat Enterprise Linux maintenance release. Product Management has 
requested further review of this request by Red Hat Engineering, for potential
inclusion in a Red Hat Enterprise Linux Update release for currently deployed 
products. This request is not yet committed for inclusion in an Update release.

Comment 4 Aristeu Rozanski 2011-03-30 14:32:19 UTC

Patch(es) available on kernel-2.6.32-128.el6

Comment 8 Eryu Guan 2011-04-06 12:33:57 UTC

The difference is not so big, but it does improve the sync performance from my testing. Set it to VERIFIED

On -71 kernel
[root@ibm-x3550m3-02 scratch]# /usr/bin/time bash -c "dd if=/dev/zero of=bigfile bs=1M count=10000 && sync"
10000+0 records in
10000+0 records out
10485760000 bytes (10 GB) copied, 90.0757 s, 116 MB/s
0.00user 22.23system 1:31.04elapsed 24%CPU (0avgtext+0avgdata 7312maxresident)k
744inputs+286130176outputs (1major+1053minor)pagefaults 0swaps

On -128 kernel
[root@ibm-x3550m3-02 scratch]# /usr/bin/time bash -c "dd if=/dev/zero of=bigfile bs=1M count=10000 && sync"
10000+0 records in
10000+0 records out
10485760000 bytes (10 GB) copied, 64.9366 s, 161 MB/s
0.00user 9.14system 1:24.04elapsed 10%CPU (0avgtext+0avgdata 7296maxresident)k
56inputs+20480000outputs (1major+1055minor)pagefaults 0swaps

Multiple runs always showed about 5 seconds improvement. 

The testing environment is a ibm-x3550 host with 16G memory and 16 cores, on a 200G ext4 partition.
[root@ibm-x3550m3-02 ~]# free -m
             total       used       free     shared    buffers     cached
Mem:         15940        441      15498          0          0         10
-/+ buffers/cache:        430      15509
Swap:            0          0          0
[root@ibm-x3550m3-02 ~]# grep -c proc /proc/cpuinfo 
16
[root@ibm-x3550m3-02 scratch]# mount | grep scratch
/dev/mapper/vg_ibmx3550m302-test on /mnt/scratch type ext4 (rw)

Comment 9 errata-xmlrpc 2011-05-23 20:43:06 UTC

An advisory has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on therefore solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.

http://rhn.redhat.com/errata/RHSA-2011-0542.html