From 41e7ccd2ce603ce479d89e73f67eaaa3707cb351 Mon Sep 17 00:00:00 2001
From: Behnam Robatmili <brobatmili@chanzuckerberg.com>
Date: Thu, 26 Oct 2023 19:52:30 -0700
Subject: [PATCH] [Python] Early version of the profiler github actions

---
 .github/workflows/profiler.yml | 21 +++++++++++++++++
 profiler/ann_data.py           | 23 +++++++++++++++++++
 profiler/perf_checker.sh       | 13 +++++++++++
 profiler/profiler.py           |  6 ++---
 profiler/top_profiler.py       | 41 ++++++++++++++++++++++++++++++++++
 5 files changed, 101 insertions(+), 3 deletions(-)
 create mode 100644 .github/workflows/profiler.yml
 create mode 100644 profiler/ann_data.py
 create mode 100755 profiler/perf_checker.sh
 create mode 100644 profiler/top_profiler.py

diff --git a/.github/workflows/profiler.yml b/.github/workflows/profiler.yml
new file mode 100644
index 0000000000..c3d8be70be
--- /dev/null
+++ b/.github/workflows/profiler.yml
@@ -0,0 +1,21 @@
+name: TileDB-SOMA Profiler run
+
+on:
+  push:
+    branches:
+  pull_request:
+    branches:
+      - "main"
+
+paths:
+      - ".github/workflows/profiler.yaml"
+jobs:
+
+  check-links:
+
+  name: Setup profiler
+
+  run: pip install ./profiler
+
+  run: |
+    ./profiler/perf_checker.sh
\ No newline at end of file
diff --git a/profiler/ann_data.py b/profiler/ann_data.py
new file mode 100644
index 0000000000..ca28e26d0b
--- /dev/null
+++ b/profiler/ann_data.py
@@ -0,0 +1,23 @@
+from time import perf_counter
+
+import cellxgene_census
+
+import tiledbsoma as soma
+
+census_S3_latest = dict(census_version="latest")
+census_local_copy = dict(uri="/Users/brobatmili/projects/census_data/")
+
+
+def main():
+    t1 = perf_counter()
+    with cellxgene_census.open_soma(**census_local_copy) as census:
+        with census["census_data"]["homo_sapiens"].axis_query(
+            measurement_name="RNA",
+            obs_query=soma.AxisQuery(value_filter="""tissue_general == 'eye'"""),
+        ) as query:
+            query.to_anndata(X_name="raw")
+    t2 = perf_counter()
+    print(f"End to end time {t2 - t1}")
+
+
+main()
diff --git a/profiler/perf_checker.sh b/profiler/perf_checker.sh
new file mode 100755
index 0000000000..451433645f
--- /dev/null
+++ b/profiler/perf_checker.sh
@@ -0,0 +1,13 @@
+#!/bin/bash
+
+python -m venv perf
+source perf/bin/activate
+pip install gitpython
+pip install psutil
+pip install comacore
+pip install profiler
+pip install tiledbsoma
+pip install cellxgene_census
+python -m profiler "python ann_data.py" -t gtime
+
+python ./top_profiler.py
\ No newline at end of file
diff --git a/profiler/profiler.py b/profiler/profiler.py
index 8e3ff3c749..282153219d 100644
--- a/profiler/profiler.py
+++ b/profiler/profiler.py
@@ -10,10 +10,9 @@
 from typing import Any, Dict, Optional
 
 import somacore
+from context_generator import host_context
 
 import tiledbsoma
-
-from context_generator import host_context
 from data import FileBasedProfileDB, ProfileData, ProfileDB
 
 GNU_TIME_FORMAT = (
@@ -207,5 +206,6 @@ def main():
         file=stderr,
     )
 
+
 if __name__ == "__main__":
-    main()
\ No newline at end of file
+    main()
diff --git a/profiler/top_profiler.py b/profiler/top_profiler.py
new file mode 100644
index 0000000000..58e6a6059a
--- /dev/null
+++ b/profiler/top_profiler.py
@@ -0,0 +1,41 @@
+import data
+
+threshold = 1.10  # Percent difference
+
+db = data.FileBasedProfileDB()
+actual_max_ts = 0
+dt = db.find("python ann_data.py")
+last_two = dt[-2:]
+c = 0
+
+for s in dt:
+    max_pointer = 0
+    second_max_pointer = 0
+    cc = 0
+    second_max_ts: float = 0
+    max_ts: float = 0
+    for ii in dt:
+        if ii.timestamp > max_ts:
+            second_max_pointer = second_max_pointer + 1
+            actual_second_max_ts = actual_max_ts
+            second_max_pointer = c - 1
+            max_ts = ii.timestamp
+            max_pointer = c
+        c = c + 1
+    print(f"max_ts {max_ts} second_max_ts {second_max_ts}")
+    print(s.user_time_sec)
+
+    print(f"Found indexes are {max_ts} {second_max_ts}")
+    L = [1, 2]
+    L[0] = dt[max_pointer].user_time_sec
+    L[1] = dt[second_max_pointer].user_time_sec
+    print(
+        f"The latest found timestamps {max_ts} time metric {L[0]} "
+        f" second timestamp {second_max_ts} time metric {L[1]}"
+    )
+    for i in range(0, len(dt)):
+        print(f"{i} dt[{i}].user_time_sec = {dt[i].user_time_sec} ts {dt[i].timestamp}")
+
+    if threshold * float(L[1]) < float(L[0]) or float(L[1]) > threshold * float(L[0]):
+        raise SystemExit(f"Potential performance degradation detected {L[0]} va {L[1]}")
+    print("No recent performance degradation detected")