Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

WIP set bfq scheduler as default #1138

Closed
wants to merge 1 commit into from
Closed

Conversation

hexfusion
Copy link
Contributor

@hexfusion hexfusion commented Sep 26, 2019

testing setting bfq as default vs mq-deadline

I was digging around and pretty surprised to see this.

[core@ip-10-0-167-161 test]$ cat /sys/block/xvda/queue/scheduler
[mq-deadline] kyber bfq none

ref: https://www.phoronix.com/scan.php?page=article&item=linux-50hdd-io&num=3

@openshift-ci-robot openshift-ci-robot added do-not-merge/work-in-progress Indicates that a PR should not merge because it is a work in progress. size/XS Denotes a PR that changes 0-9 lines, ignoring generated files. labels Sep 26, 2019
@cgwalters
Copy link
Member

See also systemd/systemd#13321

@hexfusion hexfusion changed the title WIP set bfq scheduler for as default WIP set bfq scheduler as default Sep 26, 2019
@hexfusion
Copy link
Contributor Author

After change

[core@ip-10-0-156-206 ~]$ cat /sys/block/xvda/queue/scheduler
mq-deadline kyber [bfq] none

@hexfusion
Copy link
Contributor Author

hexfusion commented Sep 26, 2019

I think testing this with AWS might be tricky to see anything definitive. But first look is solid.

fio-3.6
Starting 1 process
file1: Laying out IO file (1 file / 22MiB)

file1: (groupid=0, jobs=1): err= 0: pid=9: Thu Sep 26 13:40:24 2019
  write: IOPS=1113, BW=2502KiB/s (2562kB/s)(21.0MiB/9003msec)
    clat (usec): min=3, max=177, avg=10.89, stdev= 5.46
     lat (usec): min=4, max=1389, avg=12.28, stdev=14.97
    clat percentiles (usec):
     |  1.00th=[    7],  5.00th=[    7], 10.00th=[    7], 20.00th=[    8],
     | 30.00th=[    9], 40.00th=[    9], 50.00th=[   10], 60.00th=[   11],
     | 70.00th=[   12], 80.00th=[   14], 90.00th=[   17], 95.00th=[   19],
     | 99.00th=[   28], 99.50th=[   34], 99.90th=[   69], 99.95th=[   94],
     | 99.99th=[  155]
   bw (  KiB/s): min= 2259, max= 2659, per=99.98%, avg=2501.56, stdev=103.85, samples=18
   iops        : min= 1006, max= 1184, avg=1113.94, stdev=46.19, samples=18
  lat (usec)   : 4=0.01%, 10=52.42%, 20=44.67%, 50=2.71%, 100=0.15%
  lat (usec)   : 250=0.04%
  fsync/fdatasync/sync_file_range:
    sync (usec): min=386, max=18043, avg=876.70, stdev=546.62
    sync percentiles (usec):
     |  1.00th=[  437],  5.00th=[  457], 10.00th=[  474], 20.00th=[  502],
     | 30.00th=[  529], 40.00th=[  594], 50.00th=[  988], 60.00th=[ 1037],
     | 70.00th=[ 1074], 80.00th=[ 1123], 90.00th=[ 1188], 95.00th=[ 1270],
     | 99.00th=[ 1713], 99.50th=[ 3097], 99.90th=[ 7439], 99.95th=[10290],
     | 99.99th=[16450]
  cpu          : usr=1.39%, sys=6.60%, ctx=31940, majf=0, minf=13
  IO depths    : 1=200.0%, 2=0.0%, 4=0.0%, 8=0.0%, 16=0.0%, 32=0.0%, >=64=0.0%
     submit    : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
     complete  : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
     issued rwts: total=0,10029,0,0 short=10029,0,0,0 dropped=0,0,0,0
     latency   : target=0, window=0, percentile=100.00%, depth=1

Run status group 0 (all jobs):
  WRITE: bw=2502KiB/s (2562kB/s), 2502KiB/s-2502KiB/s (2562kB/s-2562kB/s), io=21.0MiB (23.1MB), run=9003-9003msec

Disk stats (read/write):
  xvda: ios=0/21367, merge=0/151, ticks=0/11185, in_queue=523, util=96.46%

@hexfusion
Copy link
Contributor Author

/test e2e-aws

@hexfusion
Copy link
Contributor Author

Tested this manually with Azure and saw no real improvements. Going to explore further...

@hexfusion
Copy link
Contributor Author

/close

@openshift-ci-robot
Copy link
Contributor

@hexfusion: Closed this PR.

In response to this:

/close

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

@hexfusion hexfusion deleted the bfq branch October 4, 2019 20:05
@openshift-ci-robot
Copy link
Contributor

@hexfusion: Failed to re-open PR: state cannot be changed. The bfq branch has been deleted.

In response to this:

/reopen

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

@hexfusion hexfusion restored the bfq branch December 13, 2019 01:59
@hexfusion
Copy link
Contributor Author

/reopen

@openshift-ci-robot
Copy link
Contributor

@hexfusion: Reopened this PR.

In response to this:

/reopen

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

@openshift-ci-robot
Copy link
Contributor

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: hexfusion
To complete the pull request process, please assign yuqi-zhang
You can assign the PR to them by writing /assign @yuqi-zhang in a comment when ready.

The full list of commands accepted by this bot can be found here.

Needs approval from an approver in each of these files:

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

@hexfusion
Copy link
Contributor Author

/retest

Signed-off-by: Sam Batschelet <sbatsche@redhat.com>
@openshift-ci-robot
Copy link
Contributor

openshift-ci-robot commented Dec 13, 2019

@hexfusion: The following tests failed, say /retest to rerun them all:

Test name Commit Details Rerun command
ci/prow/e2e-aws-op 37e59ccacc0da1b7fa8e5b3c3083eb701aa4c001 link /test e2e-aws-op
ci/prow/e2e-aws-disruptive 37e59ccacc0da1b7fa8e5b3c3083eb701aa4c001 link /test e2e-aws-disruptive

Full PR test history. Your PR dashboard. Please help us cut down on flakes by linking to an open issue when you hit one in your PR.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. I understand the commands that are listed here.

@hexfusion hexfusion closed this Feb 11, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
do-not-merge/work-in-progress Indicates that a PR should not merge because it is a work in progress. size/XS Denotes a PR that changes 0-9 lines, ignoring generated files.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

3 participants