Skip to content

Repository navigation

Feature/348 increase benchmark runs and fix fixed iteration ceiling #350

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

Merged
jathavaan merged 5 commits into from
May 26, 2026
Merged

Feature/348 increase benchmark runs and fix fixed iteration ceiling #350

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
c9e937c
#348 Increase benchmark runs, lower per-iteration timeout, raise fixe…
jathavaan May 26, 2026
0acd238
#348 Add per-experiment runs field and skip duckdb-large
jathavaan May 26, 2026
4580606
#348 Skip experiments that exceed their per-experiment runs limit
jathavaan May 26, 2026
3703aa0
#348 Replace skip: true with typed stop reasons and set logging to DEBUG
jathavaan May 26, 2026
115cada
#348 Changed logging level to INFO
jathavaan May 26, 2026
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
13 changes: 7 additions & 6 deletions benchmark_runner.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -45,14 +45,15 @@ def benchmark_runner() -> None:
run_id=run_id, benchmark_run=benchmark_run, dataset_size=dataset_size
)

if _is_skipped(script_id):
logger.info(f"Experiment '{script_id}' marked skip=true in benchmarks.yml. Recording as skipped.")
skip_reason = _get_skip_reason(script_id)
if skip_reason is not None:
logger.info(f"Experiment '{script_id}' skipped in benchmarks.yml (reason={skip_reason.value}).")
_save_run_metadata(
query_id=script_id,
run_id=run_id,
achieved_iterations=0,
failed_iterations=0,
stop_reason=StopReason.FAILED,
stop_reason=skip_reason,
Comment thread
jathavaan marked this conversation as resolved.
ci_half_width_seconds=None,
ci_half_width_relative=None,
mean_elapsed_seconds=None,
Expand Down Expand Up @@ -139,13 +140,13 @@ def _strip_dataset_size_suffix(script_id: str) -> str:
return script_id


def _is_skipped(script_id: str) -> bool:
def _get_skip_reason(script_id: str) -> StopReason | None:
with open(Config.BENCHMARK_FILE) as f:
cfg = yaml.safe_load(f)
for exp in cfg.get("experiments", []):
if exp.get("id") == script_id and exp.get("skip"):
return True
return False
return StopReason.from_skip(exp["skip"])
return None


def _get_args() -> tuple[str, int, Optional[str], DatasetSize]:
Expand Down
56 changes: 48 additions & 8 deletions benchmarks.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,6 +16,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: small
runs: 30
related_script_ids:
- point-in-polygon-lookup-postgis-small
- point-in-polygon-lookup-local-small
Expand All @@ -25,6 +26,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: small
runs: 30
related_script_ids:
- point-in-polygon-lookup-duckdb-small
- point-in-polygon-lookup-local-small
Expand All @@ -34,6 +36,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: small
runs: 30
related_script_ids:
- point-in-polygon-lookup-duckdb-small
- point-in-polygon-lookup-postgis-small
Expand All @@ -43,6 +46,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: large
runs: 30
related_script_ids:
- point-in-polygon-lookup-postgis-large

Expand All @@ -51,6 +55,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: large
runs: 30
related_script_ids:
- point-in-polygon-lookup-duckdb-large

Expand All @@ -60,6 +65,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: small
runs: 30
related_script_ids:
- knn-search-postgis-small
- knn-search-local-small
Expand All @@ -69,6 +75,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: small
runs: 30
related_script_ids:
- knn-search-duckdb-small
- knn-search-local-small
Expand All @@ -78,6 +85,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: small
runs: 30
related_script_ids:
- knn-search-duckdb-small
- knn-search-postgis-small
Expand All @@ -87,6 +95,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: large
runs: 30
related_script_ids:
- knn-search-postgis-large

Expand All @@ -95,6 +104,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: large
runs: 30
related_script_ids:
- knn-search-duckdb-large

Expand All @@ -104,6 +114,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: small
runs: 30
related_script_ids:
- bbox-filtering-postgis-small
- bbox-filtering-local-small
Expand All @@ -113,6 +124,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: small
runs: 30
related_script_ids:
- bbox-filtering-duckdb-small
- bbox-filtering-local-small
Expand All @@ -122,6 +134,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: small
runs: 30
related_script_ids:
- bbox-filtering-duckdb-small
- bbox-filtering-postgis-small
Expand All @@ -131,6 +144,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: large
runs: 30
related_script_ids:
- bbox-filtering-postgis-large

Expand All @@ -139,6 +153,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: large
runs: 30
related_script_ids:
- bbox-filtering-duckdb-large

Expand All @@ -155,6 +170,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: small
runs: 3
related_script_ids:
- national-scale-spatial-join-databricks-broadcast-8-nodes-small
- national-scale-spatial-join-databricks-partitioned-8-nodes-small
Expand All @@ -165,6 +181,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: small
runs: 3
related_script_ids:
- national-scale-spatial-join-databricks-broadcast-8-nodes-small
- national-scale-spatial-join-databricks-partitioned-8-nodes-small
Expand All @@ -175,6 +192,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: medium
runs: 3
related_script_ids:
- national-scale-spatial-join-databricks-broadcast-8-nodes-medium
- national-scale-spatial-join-databricks-partitioned-8-nodes-medium
Expand All @@ -185,6 +203,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: medium
runs: 3
related_script_ids:
- national-scale-spatial-join-databricks-broadcast-8-nodes-medium
- national-scale-spatial-join-databricks-partitioned-8-nodes-medium
Expand All @@ -195,6 +214,8 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: large
runs: 3
skip: timeout # exceeds 60-min per-iteration threshold; existing data from run 2026-05-24-HBJYYT sufficient
related_script_ids:
- national-scale-spatial-join-postgis-large

Expand All @@ -203,7 +224,8 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: large
skip: true # OOM on Azure PostgreSQL Flex Server at large scale
runs: 3
skip: failed # OOM on Azure PostgreSQL Flex Server at large scale
related_script_ids:
- national-scale-spatial-join-duckdb-large

Expand All @@ -213,6 +235,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: medium
runs: 3
related_script_ids:
- national-scale-spatial-join-databricks-partitioned-2-nodes-medium

Expand All @@ -221,6 +244,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: large
runs: 3
related_script_ids:
- national-scale-spatial-join-databricks-broadcast-4-nodes-large
- national-scale-spatial-join-databricks-broadcast-16-nodes-large
Expand All @@ -230,6 +254,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: small
runs: 3
related_script_ids:
- national-scale-spatial-join-databricks-partitioned-4-nodes-small

Expand All @@ -238,6 +263,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: medium
runs: 3
related_script_ids:
- national-scale-spatial-join-databricks-partitioned-4-nodes-medium

Expand All @@ -246,6 +272,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: large
runs: 3
related_script_ids:
- national-scale-spatial-join-databricks-broadcast-2-nodes-large
- national-scale-spatial-join-databricks-broadcast-16-nodes-large
Expand All @@ -255,6 +282,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: small
runs: 3
related_script_ids:
- national-scale-spatial-join-databricks-partitioned-8-nodes-small
- national-scale-spatial-join-duckdb-small
Expand All @@ -265,6 +293,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: medium
runs: 3
related_script_ids:
- national-scale-spatial-join-databricks-partitioned-8-nodes-medium
- national-scale-spatial-join-duckdb-medium
Expand All @@ -275,6 +304,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: large
runs: 3
related_script_ids:
- national-scale-spatial-join-databricks-partitioned-8-nodes-large

Expand All @@ -283,6 +313,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: large
runs: 3
related_script_ids:
- national-scale-spatial-join-databricks-broadcast-2-nodes-large
- national-scale-spatial-join-databricks-broadcast-4-nodes-large
Expand All @@ -293,7 +324,8 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: medium
skip: true # executor OOM: RangeJoin spatial index exceeds Standard_D4s_v3 memory at medium scale
runs: 3
skip: failed # executor OOM: RangeJoin spatial index exceeds Standard_D4s_v3 memory at medium scale
related_script_ids:
- national-scale-spatial-join-databricks-broadcast-2-nodes-medium

Expand All @@ -302,7 +334,8 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: large
skip: true # executor OOM: RangeJoin spatial index exceeds Standard_D4s_v3 memory at large scale
runs: 3
skip: failed # executor OOM: RangeJoin spatial index exceeds Standard_D4s_v3 memory at large scale
related_script_ids:
- national-scale-spatial-join-databricks-partitioned-4-nodes-large
- national-scale-spatial-join-databricks-partitioned-16-nodes-large
Expand All @@ -312,6 +345,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: small
runs: 3
related_script_ids:
- national-scale-spatial-join-databricks-broadcast-4-nodes-small

Expand All @@ -320,7 +354,8 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: medium
skip: true # executor OOM: RangeJoin spatial index exceeds Standard_D4s_v3 memory at medium scale
runs: 3
skip: failed # executor OOM: RangeJoin spatial index exceeds Standard_D4s_v3 memory at medium scale
related_script_ids:
- national-scale-spatial-join-databricks-broadcast-4-nodes-medium

Expand All @@ -329,7 +364,8 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: large
skip: true # executor OOM: RangeJoin spatial index exceeds Standard_D4s_v3 memory at large scale
runs: 3
skip: failed # executor OOM: RangeJoin spatial index exceeds Standard_D4s_v3 memory at large scale
related_script_ids:
- national-scale-spatial-join-databricks-partitioned-2-nodes-large
- national-scale-spatial-join-databricks-partitioned-16-nodes-large
Expand All @@ -339,6 +375,7 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: small
runs: 3
related_script_ids:
- national-scale-spatial-join-databricks-broadcast-8-nodes-small
- national-scale-spatial-join-duckdb-small
Expand All @@ -349,7 +386,8 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: medium
skip: true # executor OOM: RangeJoin spatial index exceeds Standard_D4s_v3 memory at medium scale
runs: 3
skip: failed # executor OOM: RangeJoin spatial index exceeds Standard_D4s_v3 memory at medium scale
related_script_ids:
- national-scale-spatial-join-databricks-broadcast-8-nodes-medium
- national-scale-spatial-join-duckdb-medium
Expand All @@ -360,7 +398,8 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: large
skip: true # executor OOM: RangeJoin spatial index exceeds Standard_D4s_v3 memory at large scale
runs: 3
skip: failed # executor OOM: RangeJoin spatial index exceeds Standard_D4s_v3 memory at large scale
related_script_ids:
- national-scale-spatial-join-databricks-broadcast-8-nodes-large

Expand All @@ -369,7 +408,8 @@ experiments:
cpu: 4
memory_gb: 16
dataset_size: large
skip: true # executor OOM: RangeJoin spatial index exceeds Standard_D4s_v3 memory at large scale
runs: 3
skip: failed # executor OOM: RangeJoin spatial index exceeds Standard_D4s_v3 memory at large scale
related_script_ids:
- national-scale-spatial-join-databricks-partitioned-2-nodes-large
- national-scale-spatial-join-databricks-partitioned-4-nodes-large
Expand Down
11 changes: 11 additions & 0 deletions main.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -74,6 +74,17 @@ def _run_benchmarks(
if experiment_id in completed_experiments:
continue

experiment_runs = experiment.get("runs", Config.BENCHMARK_RUNS)
if benchmark_run > experiment_runs:
completed_experiments.append(experiment_id)
for rid in experiment.get("related_script_ids", []):
completed_experiments.append(str(rid))
logger.info(
"Skipping '%s' — benchmark run %s exceeds experiment runs limit %s.",
experiment_id, benchmark_run, experiment_runs,
)
Comment thread
jathavaan marked this conversation as resolved.
continue

current_batch += 1

related_experiment_ids = experiment["related_script_ids"]
Expand Down
Loading
Loading