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.

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

Jump to bottom

Beaker workspace performance #328

Merged
merged 3 commits into from
Jul 5, 2022
Merged

Beaker workspace performance #328

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
e4b1271
Improve performance of `BeakerWorkspace`
epwalsh Jun 27, 2022
fa392c0
bump beaker-py
epwalsh Jun 27, 2022
d1d1543
Merge branch 'main' into beaker-workspace-perf
dirkgr Jul 1, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
1 change: 1 addition & 0 deletions CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,6 +12,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
- Improved `Step.ensure_result()` such that the step's result doesn't have to be read from the cache.
- Fixed an issue with the output from `MulticoreExecutor` such that it's now consistent with the default `Executor` for steps that were found in the cache.
- One of our error messages referred to a configuration file that no longer exists.
- Improved performance of `BeakerWorkspace`.


## [v0.9.1](https://github.com/allenai/tango/releases/tag/v0.9.1) - 2022-06-24
Expand Down
2 changes: 1 addition & 1 deletion requirements.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,7 +35,7 @@ pytorch-lightning>=1.6,<1.7 # needed by: pytorch_lightning
transformers>=4.12.3 # needed by: transformers
sentencepiece>=0.1.96 # needed by: transformers
fairscale==0.4.6 # needed by: fairscale
beaker-py>=1.3.0,<2.0.0 # needed by: beaker
beaker-py>=1.6.2,<2.0.0 # needed by: beaker

# sacremoses should be a dependency of transformers, but it is missing, so we add it manually.
sacremoses # needed by: transformers
Expand Down
6 changes: 3 additions & 3 deletions tango/integrations/beaker/workspace.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -46,7 +46,7 @@ class BeakerWorkspace(Workspace):

def __init__(self, workspace: str, **kwargs):
super().__init__()
self.beaker = Beaker.from_env(default_workspace=workspace, **kwargs)
self.beaker = Beaker.from_env(default_workspace=workspace, session=True, **kwargs)
self.cache = BeakerStepCache(beaker=self.beaker)
self.steps_dir = tango_cache_dir() / "beaker_workspace"
self.locks: Dict[Step, BeakerStepLock] = {}
Expand Down Expand Up @@ -233,7 +233,7 @@ def registered_runs(self) -> Dict[str, Run]:
runs: Dict[str, Run] = {}

with concurrent.futures.ThreadPoolExecutor(
max_workers=3, thread_name_prefix="BeakerWorkspace.registered_runs()-"
max_workers=9, thread_name_prefix="BeakerWorkspace.registered_runs()-"
) as executor:
run_futures = []
for dataset in self.beaker.workspace.datasets(
Expand Down Expand Up @@ -295,7 +295,7 @@ def _get_run_from_dataset(self, dataset: Dataset) -> Optional[Run]:
import concurrent.futures

with concurrent.futures.ThreadPoolExecutor(
max_workers=3, thread_name_prefix="BeakerWorkspace._get_run_from_dataset()-"
max_workers=9, thread_name_prefix="BeakerWorkspace._get_run_from_dataset()-"
) as executor:
step_info_futures = []
for step_name, unique_id in steps_info.items():
Expand Down