Skip to content

Adding more parameters to the sweeper combinations for FUSE threads and objects. #1592

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.

Sign up for GitHub

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

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Adding more parameters to the sweeper combinations for FUSE threads and objects. #1592

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion benchmark/conf/hydra/sweeper/base.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,7 @@
_target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
max_batch_size: null
params:
'application_workers': 1, 16 # 4, 64
'application_workers': 1, 4, 16, 64
'iteration': "range(${iterations})"
'network.maximum_throughput_gbps': 400
# 'crt_eventloop_threads': 128, 256, 1024
2 changes: 1 addition & 1 deletion benchmark/conf/hydra/sweeper/fio.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,4 +1,4 @@
# @package hydra.sweeper
params:
'benchmarks.fio.direct_io': false, true
'benchmarks.fio.fuse_threads': 16 #1, 64
'benchmarks.fio.fuse_threads': 1, 16, 64
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It'd be good to restrict the combinations of app workers and fuse_threads as we don't care about all combinations of these values. Can we specify them as a pair instead?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I had the same suggestion at some point too, but it makes me wonder how feasible it is to maintain and extend a coupled setup like that.

  • Does it become more error-prone or something that requires context to update every time?
  • Do we really need an increased number of combinations and consequently a stricter sweeper (in mainline), given we only plot 1 object and 16 objects on day-to-day basis; or should this be something one can play around with during investigations on their machine?
  • Will our Jupyter notebooks continue to work with some values/combinations of threads/app_workers missing?

Loading