Skip to content

Bug 2021696 - Use a fixed chunk size in the Perfherder data cycler #9533

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
junngo wants to merge 1 commit into
base: master
Choose a base branch
from
+8 −28
Open

Bug 2021696 - Use a fixed chunk size in the Perfherder data cycler #9533

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
36 changes: 8 additions & 28 deletions treeherder/model/data_cycling/removal_strategies.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -84,7 +84,6 @@ def remove(self, using: CursorWrapper):
Although the WHERE clause in del_replicate looks redundant, it is intentionally kept to guide
the PostgreSQL planner toward a more efficient execution plan.
"""
chunk_size = self._find_ideal_chunk_size()
using.execute(
"""
WITH target_datum AS (
Expand Down Expand Up @@ -116,21 +115,13 @@ def remove(self, using: CursorWrapper):
USING target_datum td
WHERE pd.id = td.id
""",
[self._max_timestamp, chunk_size, self._max_timestamp],
[self._max_timestamp, self._chunk_size, self._max_timestamp],
)

@property
def name(self) -> str:
return "main removal strategy"

def _find_ideal_chunk_size(self) -> int:
max_id = self._manager.filter(push_timestamp__gt=self._max_timestamp).order_by("-id")[0].id
older_ids = self._manager.filter(
push_timestamp__lte=self._max_timestamp, id__lte=max_id
).order_by("id")[: self._chunk_size]

return len(older_ids) or self._chunk_size


class TryDataRemoval(RemovalStrategy):
"""
Expand Down Expand Up @@ -323,7 +314,6 @@ def remove(self, using: CursorWrapper):
Although the WHERE clause in del_replicate looks redundant, it is intentionally kept to guide
the PostgreSQL planner toward a more efficient execution plan.
"""
chunk_size = self._find_ideal_chunk_size()
repository_id = self.irrelevant_repo
using.execute(
"""
Expand Down Expand Up @@ -357,24 +347,14 @@ def remove(self, using: CursorWrapper):
USING target_datum td
WHERE pd.id = td.id
""",
[repository_id, self._max_timestamp, chunk_size, repository_id, self._max_timestamp],
)

def _find_ideal_chunk_size(self) -> int:
max_id_of_non_expired_row = (
self._manager.filter(push_timestamp__gt=self._max_timestamp)
.filter(repository_id__in=self.irrelevant_repositories)
.order_by("-id")[0]
.id
)
older_perf_data_rows = (
self._manager.filter(
push_timestamp__lte=self._max_timestamp, id__lte=max_id_of_non_expired_row
)
.filter(repository_id__in=self.irrelevant_repositories)
.order_by("id")[: self._chunk_size]
[
repository_id,
self._max_timestamp,
self._chunk_size,
repository_id,
self._max_timestamp,
],
)
return len(older_perf_data_rows) or self._chunk_size


class StalledDataRemoval(RemovalStrategy):
Expand Down