Bug 1429231 - [Perf] : SMBv1 Sequential Writes are off target by 20% on plain distribute volumes.
Summary: [Perf] : SMBv1 Sequential Writes are off target by 20% on plain distribute vo...
Keywords:
Status: CLOSED INSUFFICIENT_DATA
Alias: None
Product: Red Hat Gluster Storage
Classification: Red Hat Storage
Component: io-threads
Version: rhgs-3.2
Hardware: x86_64
OS: Linux
unspecified
high
Target Milestone: ---
: ---
Assignee: Raghavendra G
QA Contact: Nag Pavan Chilakam
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2017-03-05 18:41 UTC by Ambarish
Modified: 2018-11-12 04:45 UTC (History)
6 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2018-11-12 04:45:20 UTC
Embargoed:


Attachments (Terms of Use)
313_Seq Writes (13.86 KB, application/vnd.oasis.opendocument.text)
2017-03-06 11:02 UTC, Ambarish
no flags Details


Links
System ID Private Priority Status Summary Last Updated
Red Hat Bugzilla 1397854 0 unspecified CLOSED [Perf] 10% and 20% drop in sequential writes on SMB v1 and V3 with RHEL 6.8 with a *2 deployment 2023-09-14 03:34:55 UTC

Internal Links: 1397854

Description Ambarish 2017-03-05 18:41:50 UTC
Description of problem:
-----------------------

A regression on SMBv1 seq writes was introduced in latest 3.2 bits :

3.1.3 : 1716886 kB/sec

3.2 : 1388869 kB/sec

Regression : ~20%

Version-Release number of selected component (if applicable):
-------------------------------------------------------------

3.8.4-15

How reproducible:
-----------------

Every time.



Actual results:
---------------

20% regression on SMBv1 seq writes with io-threads "on" on 3.2 bits.


Expected results:
----------------

Regression threshold : 10%

Additional info:
----------------

Volume Name: testvol
Type: Distribute
Volume ID: 35b73a47-bdc7-48b2-81a1-9b66624ae57c
Status: Started
Snapshot Count: 0
Number of Bricks: 4
Transport-type: tcp
Bricks:
Brick1: gqas014.sbu.lab.eng.bos.redhat.com:/bricks/testvol_brick0
Brick2: gqas005.sbu.lab.eng.bos.redhat.com:/bricks/testvol_brick1
Brick3: gqas006.sbu.lab.eng.bos.redhat.com:/bricks/testvol_brick2
Brick4: gqas015.sbu.lab.eng.bos.redhat.com:/bricks/testvol_brick3
Options Reconfigured:
network.inode-lru-limit: 90000
performance.md-cache-timeout: 600
performance.cache-invalidation: on
performance.cache-samba-metadata: on
performance.stat-prefetch: on
features.cache-invalidation-timeout: 600
features.cache-invalidation: on
client.event-threads: 2
server.event-threads: 2
cluster.lookup-optimize: off
performance.client-io-threads: on
transport.address-family: inet
performance.readdir-ahead: on
nfs.disable: off

Comment 3 Ambarish 2017-03-05 18:50:53 UTC
3.1.3 : 1716886 kB/sec

3.2 Defaults : 1488869 kB/sec

3.2 io-threads off : 1673075.3 kB/sec


Switching off io-threads brings back the lost regression.

Comment 7 Ambarish 2017-03-06 11:02:14 UTC
Created attachment 1260362 [details]
313_Seq Writes

Comment 14 Atin Mukherjee 2018-11-09 03:35:06 UTC
Is this still an issue? If not can we close this bug?

Comment 15 Raghavendra G 2018-11-09 04:01:39 UTC
(In reply to Atin Mukherjee from comment #14)
> Is this still an issue? If not can we close this bug?

I don't have much data to answer the question. Since the regression might've carried out over to next releases, we may not have seen this between 3.3 and 3.4 regression suites.

Comment 16 Atin Mukherjee 2018-11-09 13:37:03 UTC
In that case please close the bug.

Comment 17 Raghavendra G 2018-11-12 04:45:20 UTC
Based on comment #16


Note You need to log in before you can comment on or make changes to this bug.