Search String: Display: Description: Sort:

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*XFS\s+performance\s+tracking\s+and\s+regression\s+monitoring\s*$/: 12 ]

Total 12 documents matching your query.

1. in xfs_bulkstat (score: 1)
Author: xxxxxxxxxxx>
Date: Fri, 24 Oct 2008 09:29:42 +1000
We're about to deploy a system+jbod dedicated for performance regression tracking. The idea is to build the XFS dev branch nightly, run a bunch of self contained benchmarks, and generate a progressiv
/archives/xfs/2008-10/msg00516.html (8,491 bytes)

2. deadlock with latest xfs (score: 1)
Author: xxxxxxxxxxx>
Date: Fri, 24 Oct 2008 14:54:11 +1100
wallclock runtime is not indicative of relative performance for many benchmarks. e.g. dbench runs for a fixed time and then gives a throughput number as it's output. It's the throughput you want to c
/archives/xfs/2008-10/msg00523.html (10,045 bytes)

3. (score: 1)
Author: xxxxxxxxxxx>
Date: Fri, 24 Oct 2008 17:12:13 +1000
Dave Chinner wrote: On Fri, Oct 24, 2008 at 09:29:42AM +1000, Mark Goodwin wrote: We're about to deploy a system+jbod dedicated for performance regression tracking. The idea is to build the XFS dev b
/archives/xfs/2008-10/msg00530.html (12,306 bytes)

4. and regression monitoring (score: 1)
Author: xxxxxxxxxxx>
Date: Fri, 24 Oct 2008 10:37:58 -0500
I was going to suggest that too, those are some nifty charts. :) ffsb takes workload recipes so you can make it do a large variety of things... -Eric
/archives/xfs/2008-10/msg00537.html (8,129 bytes)

5. sometimes) faster (score: 1)
Author: ve Chinner <david@xxxxxxxxxxxxx>
Date: Fri, 24 Oct 2008 09:29:42 +1000
We're about to deploy a system+jbod dedicated for performance regression tracking. The idea is to build the XFS dev branch nightly, run a bunch of self contained benchmarks, and generate a progressiv
/archives/xfs/2008-10/msg01337.html (8,308 bytes)

6. odes using destroy_inode (score: 1)
Author: ic Sandeen <sandeen@xxxxxxxxxxx>
Date: Fri, 24 Oct 2008 14:54:11 +1100
wallclock runtime is not indicative of relative performance for many benchmarks. e.g. dbench runs for a fixed time and then gives a throughput number as it's output. It's the throughput you want to c
/archives/xfs/2008-10/msg01344.html (9,866 bytes)

7. ith latest xfs (score: 1)
Author: ve Chinner <david@xxxxxxxxxxxxx>
Date: Fri, 24 Oct 2008 17:12:13 +1000
Dave Chinner wrote: On Fri, Oct 24, 2008 at 09:29:42AM +1000, Mark Goodwin wrote: We're about to deploy a system+jbod dedicated for performance regression tracking. The idea is to build the XFS dev b
/archives/xfs/2008-10/msg01351.html (12,123 bytes)

8. fs (score: 1)
Author: tem <MAILER-DAEMON@xxxxxxxxxxxx>
Date: Fri, 24 Oct 2008 10:37:58 -0500
I was going to suggest that too, those are some nifty charts. :) ffsb takes workload recipes so you can make it do a large variety of things... -Eric
/archives/xfs/2008-10/msg01358.html (7,950 bytes)

9. XFS performance tracking and regression monitoring (score: 1)
Author: Mark Goodwin <markgw@xxxxxxx>
Date: Fri, 24 Oct 2008 09:29:42 +1000
We're about to deploy a system+jbod dedicated for performance regression tracking. The idea is to build the XFS dev branch nightly, run a bunch of self contained benchmarks, and generate a progressiv
/archives/xfs/2008-10/msg02155.html (8,128 bytes)

10. Re: XFS performance tracking and regression monitoring (score: 1)
Author: Dave Chinner <david@xxxxxxxxxxxxx>
Date: Fri, 24 Oct 2008 14:54:11 +1100
wallclock runtime is not indicative of relative performance for many benchmarks. e.g. dbench runs for a fixed time and then gives a throughput number as it's output. It's the throughput you want to c
/archives/xfs/2008-10/msg02162.html (9,929 bytes)

11. Re: XFS performance tracking and regression monitoring (score: 1)
Author: Mark Goodwin <markgw@xxxxxxx>
Date: Fri, 24 Oct 2008 17:12:13 +1000
Dave Chinner wrote: On Fri, Oct 24, 2008 at 09:29:42AM +1000, Mark Goodwin wrote: We're about to deploy a system+jbod dedicated for performance regression tracking. The idea is to build the XFS dev b
/archives/xfs/2008-10/msg02169.html (12,163 bytes)

12. Re: XFS performance tracking and regression monitoring (score: 1)
Author: Eric Sandeen <sandeen@xxxxxxxxxxx>
Date: Fri, 24 Oct 2008 10:37:58 -0500
I was going to suggest that too, those are some nifty charts. :) ffsb takes workload recipes so you can make it do a large variety of things... -Eric
/archives/xfs/2008-10/msg02176.html (8,105 bytes)


This search system is powered by Namazu