feat: tpch + tpcds GHA launcher

benchmarking/tpch/ray_job_runner.py

@@ -1,3 +1,11 @@
+# /// script
+# requires-python = ">=3.12"
+# dependencies = [
+#   "ray[default]",
+#   "getdaft",
+# ]
+# ///
+
 from __future__ import annotations
 
 import argparse

tools/tpcds.py

@@ -0,0 +1,70 @@
+# /// script
+# requires-python = ">=3.12"
+# dependencies = [
+#   "PyGithub",
+#   "boto3",
+# ]
+# ///
+
+import argparse
+import json
+
+import github
+import utils
+
+
+def run(
+    branch_name: str,
+    questions: str,
+    scale_factor: int,
+    cluster_profile: str,
+    env_vars: str,
+):
+    branch_name, _ = utils.get_name_and_commit_hash(branch_name)
+
+    expanded_questions = utils.parse_questions(questions, 99)
+    print(f"Running scale-factor of {scale_factor}GB on questions: {', '.join(map(str, expanded_questions))}")
+    args_as_list = [f"--question={q} --scale-factor={scale_factor}" for q in expanded_questions]
+    entrypoint_args = json.dumps(args_as_list)
+
+    workflow = utils.repo.get_workflow("run-cluster.yaml")
+    utils.dispatch(
+        workflow=workflow,
+        branch_name=branch_name,
+        inputs={
+            "cluster_profile": cluster_profile,
+            "working_dir": "benchmarking/tpcds",
+            "entrypoint_script": "ray_entrypoint.py",
+            "entrypoint_args": entrypoint_args,
+            "env_vars": env_vars,
+        },
+    )
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--ref", type=str, required=False, help="The branch name to run on")
+    parser.add_argument(
+        "--questions", type=str, required=False, default="*", help="A comma separated list of questions to run"
+    )
+    parser.add_argument("--scale-factor", type=int, required=False, default=2, help="The scale factor to run on")
+    parser.add_argument("--cluster-profile", type=str, required=False, help="The ray cluster configuration to run on")
+    parser.add_argument(
+        "--env-vars",
+        type=str,
+        required=False,
+        help="A comma separated list of environment variables to pass to ray job",
+    )
+    parser.add_argument("--verbose", action="store_true", help="Verbose debugging")
+    args = parser.parse_args()
+
+    if args.verbose:
+        github.enable_console_debug_logging()
+
+    run(
+        branch_name=args.ref,
+        questions=args.questions,
+        scale_factor=args.scale_factor,
+        cluster_profile=args.cluster_profile,
+        env_vars=args.env_vars,
+    )

tools/tpch.py

@@ -0,0 +1,80 @@
+# /// script
+# requires-python = ">=3.12"
+# dependencies = [
+#   "PyGithub",
+#   "boto3",
+# ]
+# ///
+
+import argparse
+import json
+
+import github
+import utils
+
+TOTAL_NUMBER_OF_QUESTIONS = 22
+
+
+def run(
+    branch_name: str,
+    questions: str,
+    scale_factor: int,
+    num_partitions: int,
+    cluster_profile: str,
+    env_vars: str,
+):
+    branch_name, _ = utils.get_name_and_commit_hash(branch_name)
+
+    expanded_questions = utils.parse_questions(questions, TOTAL_NUMBER_OF_QUESTIONS)
+    print(
+        f"Running scale-factor of {scale_factor}GB with {num_partitions} partitions on questions: {', '.join(map(str, expanded_questions))}"
+    )
+    args_as_list = [
+        f"--question-number={q} --parquet-folder=s3://eventual-dev-benchmarking-fixtures/uncompressed/tpch-dbgen/{scale_factor}_0/{num_partitions}/parquet/"
+        for q in expanded_questions
+    ]
+    entrypoint_args = json.dumps(args_as_list)
+
+    workflow = utils.repo.get_workflow("run-cluster.yaml")
+    utils.dispatch(
+        workflow=workflow,
+        branch_name=branch_name,
+        inputs={
+            "cluster_profile": cluster_profile,
+            "working_dir": "benchmarking/tpch",
+            "entrypoint_script": "ray_job_runner.py",
+            "entrypoint_args": entrypoint_args,
+            "env_vars": env_vars,
+        },
+    )
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--ref", type=str, required=False, help="The branch name to run on")
+    parser.add_argument(
+        "--questions", type=str, required=False, default="*", help="A comma separated list of questions to run"
+    )
+    parser.add_argument("--scale-factor", type=int, required=False, default=2, help="The scale factor to run on")
+    parser.add_argument("--num-partitions", type=int, required=False, default=2, help="The number of partitions")
+    parser.add_argument("--cluster-profile", type=str, required=False, help="The ray cluster configuration to run on")
+    parser.add_argument(
+        "--env-vars",
+        type=str,
+        required=False,
+        help="A comma separated list of environment variables to pass to ray job",
+    )
+    parser.add_argument("--verbose", action="store_true", help="Verbose debugging")
+    args = parser.parse_args()
+
+    if args.verbose:
+        github.enable_console_debug_logging()
+
+    run(
+        branch_name=args.ref,
+        questions=args.questions,
+        scale_factor=args.scale_factor,
+        num_partitions=args.num_partitions,
+        cluster_profile=args.cluster_profile,
+        env_vars=args.env_vars,
+    )

tools/utils.py

@@ -0,0 +1,116 @@
+import subprocess
+import time
+import typing
+from typing import Optional
+
+import gha_run_cluster_job
+from github import Auth, Github
+from github.Workflow import Workflow
+from github.WorkflowRun import WorkflowRun
+
+RETRY_ATTEMPTS = 5
+
+auth = Auth.Token(gha_run_cluster_job.get_oauth_token())
+g = Github(auth=auth)
+repo = g.get_repo("Eventual-Inc/Daft")
+
+
+def dispatch(workflow: Workflow, branch_name: str, inputs: dict) -> WorkflowRun:
+    pre_creation_latest_run = get_latest_run(workflow)
+
+    print(f"Launching workflow '{workflow.name}' on the branch '{branch_name}'")
+    created = workflow.create_dispatch(
+        ref=branch_name,
+        inputs=inputs,
+    )
+    if not created:
+        raise RuntimeError("Could not create workflow, suggestion: run again with --verbose")
+
+    post_creation_latest_run = None
+    for _ in range(RETRY_ATTEMPTS):
+        post_creation_latest_run = get_latest_run(workflow)
+        if pre_creation_latest_run.run_number == post_creation_latest_run.run_number:
+            sleep_and_then_retry()
+        elif pre_creation_latest_run.run_number < post_creation_latest_run.run_number:
+            break
+        else:
+            typing.assert_never(
+                "Run numbers are always returned in sorted order and are always monotonically increasing"
+            )
+    if not post_creation_latest_run:
+        raise RuntimeError(f"Unable to locate the new run request for the '{workflow.name}' workflow")
+
+    print(f"Launched new '{workflow.name}' workflow with id: {post_creation_latest_run.id}")
+    print(f"View the workflow run at: {post_creation_latest_run.html_url}")
+
+    return post_creation_latest_run
+
+
+def sleep_and_then_retry(sleep_amount_sec: int = 3):
+    time.sleep(sleep_amount_sec)
+
+
+def get_latest_run(workflow: Workflow) -> WorkflowRun:
+    for _ in range(RETRY_ATTEMPTS):
+        runs = workflow.get_runs()
+
+        if runs.totalCount > 0:
+            return runs[0]
+
+        sleep_and_then_retry()
+
+    raise RuntimeError("Unable to list all workflow invocations")
+
+
+def get_name_and_commit_hash(branch_name: Optional[str]) -> tuple[str, str]:
+    branch_name = branch_name or "HEAD"
+    name = (
+        subprocess.check_output(["git", "rev-parse", "--abbrev-ref", branch_name], stderr=subprocess.STDOUT)
+        .strip()
+        .decode("utf-8")
+    )
+    commit_hash = (
+        subprocess.check_output(["git", "rev-parse", branch_name], stderr=subprocess.STDOUT).strip().decode("utf-8")
+    )
+    return name, commit_hash
+
+
+def parse_questions(questions: str, total_number_of_questions: int) -> list[int]:
+    if not questions:
+        return []
+
+    if questions == "*":
+        return list(range(1, total_number_of_questions + 1))
+
+    items = questions.split(",")
+    nums = []
+    for item in items:
+        try:
+            num = int(item)
+            if num > total_number_of_questions:
+                raise RuntimeError(
+                    f"Requested question number ({num}) is greater than the total number of questions available ({total_number_of_questions})"
+                )
+            nums.append(str(num))
+            continue
+        except ValueError:
+            ...
+
+        if "-" not in item:
+            raise ValueError("...")
+
+        try:
+            lower, upper = item.split("-")
+            lower_int = int(lower)
+            upper_int = int(upper)
+            if lower_int > upper_int:
+                raise ValueError
+            if upper_int > total_number_of_questions:
+                raise RuntimeError(
+                    f"Requested question number ({upper_int}) is greater than the total number of questions available ({total_number_of_questions})"
+                )
+            nums.extend(range(lower_int, upper_int + 1))
+        except ValueError:
+            raise ValueError(f"Invalid question item; expected a number or a range, instead got {item}")
+
+    return nums