feature(metrics): dump stat to a file

tests/unittests/test_dump_metrics.py

@@ -0,0 +1,81 @@
+# Copyright 2025 Rebellions Inc. All rights reserved.
+
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+
+#     http://www.apache.org/licenses/LICENSE-2.0
+
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""This writes metrics dump files when enabled.
+
+Requires RBLN NPU, network (HF model), and opt-in:
+  VLLM_RBLN_DUMP_METRICS=1 pytest ...
+"""
+
+from __future__ import annotations
+
+import os
+import subprocess
+import sys
+from pathlib import Path
+
+import pytest
+
+# Smallest public Qwen2 causal LM in the Qwen2 family (~0.5B params).
+QWEN2_SMALL_MODEL = "Qwen/Qwen3-0.6B"
+
+REPO_ROOT = Path(__file__).resolve().parents[2]
+OFFLINE_INFERENCE_BASIC = REPO_ROOT / "examples" / "experimental" / "offline_inference_basic.py"
+
+
+pytestmark = pytest.mark.skipif(
+    os.environ.get("VLLM_RBLN_DUMP_METRICS", "").lower()
+    not in ("1", "true", "yes"),
+    reason="Set VLLM_RBLN_DUMP_METRICS=1 to run (needs RBLN + HF).",
+)
+
+
+def test_dump_metrics(tmp_path: Path) -> None:
+    """Run dump metrics; expect *_metrics.txt in cwd."""
+    env = os.environ.copy()
+    env["RBLN_USE_CUSTOM_KERNEL"] = "0"
+    env["VLLM_RBLN_COMPILE_STRICT_MODE"] = "1"
+    env["VLLM_DISABLE_COMPILE_CACHE"] = "1"
+    env["VLLM_USE_V1"] = "1"
+    env["VLLM_RBLN_METRICS"] = "1"
+    env["VLLM_RBLN_DUMP_METRICS"] = "1"
+
+    cmd = [
+        sys.executable,
+        str(OFFLINE_INFERENCE_BASIC),
+        "--model",
+        QWEN2_SMALL_MODEL,
+        "--max-num-seqs",
+        "1",
+        "--block-size",
+        "4096",
+        "--tensor-parallel-size",
+        "1",
+    ]
+    subprocess.run(
+        cmd,
+        cwd=tmp_path,
+        env=env,
+        check=True,
+        timeout=7200,
+    )
+
+    dumped = list(tmp_path.glob("*_metrics.txt"))
+    assert dumped, (
+        f"Expected at least one *_metrics.txt under {tmp_path}, "
+        f"got {list(tmp_path.iterdir())}"
+    )
+    for path in dumped:
+        text = path.read_text()
+        assert "METRICS" in text, f"{path} should contain METRICS section, got {text!r}"

vllm_rbln/v1/worker/metrics.py

@@ -13,9 +13,11 @@
 # limitations under the License.
 
 import atexit
+import os
 from collections import defaultdict
 from dataclasses import dataclass, field
 
+import vllm_rbln.rbln_envs as envs
 from vllm_rbln.logger import init_logger
 
 logger = init_logger(__name__)
@@ -130,27 +132,37 @@
         """Get total number of requests processed."""
         return len(self.latencies)
 
-    def show_stats(self, stat_type: str):
+    def gen_stats(self, stat_type: str) -> str:
+        stats = f""
         if self.get_call_counts() > 0:
-            logger.info("%s METRICS:", stat_type)
-            logger.info("  Total call counts: %d", self.get_call_counts())
-            logger.info("  Average latency: %.2f ms", self.get_avg_latency())
+            stats += f"{stat_type} METRICS:\n"
+            stats += f"  Total call counts: {self.get_call_counts()}\n"
+            stats += f"  Average latency: {self.get_avg_latency()} ms\n"
             if sum(self.token_counts) > 0:
-                logger.info("  Total tokens processed: %d", sum(self.token_counts))
-                logger.info(
-                    "  Average throughput: %.2f tokens/sec", self.get_avg_throughput()
-                )
+                stats += f"  Total tokens processed: {sum(self.token_counts)}\n"
+                stats += f"  Average throughput: {self.get_avg_throughput()} tokens/sec\n"
             if self.host_times:
-                logger.info("  Average host time: %.2f us", self.get_avg_host_time())
+                stats += f"  Average host time: {self.get_avg_host_time()} us\n"
             if self.device_times:
-                logger.info(
-                    "  Average device time: %.2f us", self.get_avg_device_time()
-                )
+                stats += f"  Average device time: {self.get_avg_device_time()} us\n"
             if self.ccl_times:
-                logger.info("  Average ccl time: %.2f us", self.get_avg_ccl_time())
+                stats += f"  Average ccl time: {self.get_avg_ccl_time()} us\n"
         else:
-            logger.info("%s METRICS: No data recorded", stat_type)
+            stats += f"{stat_type} METRICS: No data recorded\n"
+        return stats
+
+    def dump_stats(self, stat_type: str, stats: str):
+        filename = f"{stat_type}_metrics.txt"
+        if os.path.exists(filename):
+            os.remove(filename)
+        with open(filename, "w") as f:
+            f.write(stats)
 
+    def show_stats(self, stat_type: str):
+        stats = self.gen_stats(stat_type)
+        logger.info(stats)
+        if envs.VLLM_RBLN_DUMP_METRICS:
+            self.dump_stats(stat_type, stats)
 
 class PrefillMetricsByRequestID:
     """Metrics for prefill step by request id."""