[1/2] L2 CI: End to End Performance#2884
Merged
ApostaC merged 6 commits intoLMCache:devfrom Mar 27, 2026
Merged
Conversation
Signed-off-by: Oasis-Git <ayw.sirius19@gmail.com>
Contributor
There was a problem hiding this comment.
Code Review
This pull request introduces a new performance test script, run-long-doc-qa-l2.sh, which evaluates LMCache's L2 caching capabilities by restarting the server with specific L2 configurations and verifying speedup thresholds. The main test runner, run-mp-test.sh, has been updated to include this new test step. Feedback focuses on improving script consistency by using python3, enhancing modularity by moving a large embedded Python heredoc into a standalone file, and ensuring the test suite continues executing subsequent tests even if the L2 test fails.
| GPU_DEVICE="${GPU_FOR_VLLM:-0}" | ||
|
|
||
| CUDA_VISIBLE_DEVICES="${GPU_DEVICE}" \ | ||
| python -m lmcache.v1.multiprocess.server \ |
Contributor
There was a problem hiding this comment.
For consistency and to avoid potential issues in environments where python might point to Python 2, it's better to use python3 here. The rest of the script already uses python3 (e.g., lines 129, 150, 195).
Suggested change
| python -m lmcache.v1.multiprocess.server \ | |
| python3 -m lmcache.v1.multiprocess.server \ |
Comment on lines
+195
to
+267
| python3 << EOF | ||
| import sys | ||
|
|
||
| def sf(val): | ||
| try: return float(val) | ||
| except: return None | ||
|
|
||
| bqt = sf("$baseline_query_ttft") | ||
| bqrt = sf("$baseline_query_round_time") | ||
| bwrt = sf("$baseline_warmup_round_time") | ||
| lqt = sf("$l2_query_ttft") | ||
| lqrt = sf("$l2_query_round_time") | ||
| lwrt = sf("$l2_warmup_round_time") | ||
|
|
||
| min_spd = float("$MIN_L2_SPEEDUP") | ||
| min_ttft = float("$MIN_L2_TTFT_SPEEDUP") | ||
| max_oh = float("$MAX_WARMUP_OVERHEAD") | ||
|
|
||
| failed = False | ||
|
|
||
| print("=" * 60) | ||
| print("L2 Performance Summary") | ||
| print("=" * 60) | ||
| print(f"{'Metric':<35} {'Baseline':>12} {'L2':>12}") | ||
| print("-" * 60) | ||
| for name, bv, lv in [ | ||
| ("query_ttft_per_prompt (s)", bqt, lqt), | ||
| ("query_round_time_per_prompt (s)", bqrt, lqrt), | ||
| ("warmup_round_time_per_prompt (s)", bwrt, lwrt), | ||
| ]: | ||
| bs = f"{bv:.4f}" if bv else "N/A" | ||
| ls = f"{lv:.4f}" if lv else "N/A" | ||
| print(f"{name:<35} {bs:>12} {ls:>12}") | ||
|
|
||
| print() | ||
| print("=" * 60) | ||
| print("Threshold Verification") | ||
| print("=" * 60) | ||
|
|
||
| # 1. L2 query round-time speedup | ||
| if lqrt and bqrt and lqrt > 0: | ||
| s = bqrt / lqrt | ||
| ok = s >= min_spd | ||
| print(f"[{'PASS' if ok else 'FAIL'}] L2 query speedup: {s:.2f}x (need >= {min_spd}x)") | ||
| if not ok: failed = True | ||
| else: | ||
| print("[FAIL] Cannot compute L2 query speedup"); failed = True | ||
|
|
||
| # 2. L2 TTFT speedup | ||
| if lqt and bqt and lqt > 0: | ||
| s = bqt / lqt | ||
| ok = s >= min_ttft | ||
| print(f"[{'PASS' if ok else 'FAIL'}] L2 TTFT speedup: {s:.2f}x (need >= {min_ttft}x)") | ||
| if not ok: failed = True | ||
| else: | ||
| print("[FAIL] Cannot compute L2 TTFT speedup"); failed = True | ||
|
|
||
| # 3. Warmup overhead | ||
| if lwrt and bwrt and bwrt > 0: | ||
| o = lwrt / bwrt | ||
| ok = o <= max_oh | ||
| print(f"[{'PASS' if ok else 'FAIL'}] Warmup overhead: {o:.2f}x (need <= {max_oh}x)") | ||
| if not ok: failed = True | ||
| else: | ||
| print("[FAIL] Cannot compute warmup overhead"); failed = True | ||
|
|
||
| print() | ||
| if failed: | ||
| print("[FAIL] L2 performance verification FAILED") | ||
| sys.exit(1) | ||
| else: | ||
| print("[PASS] All L2 performance thresholds passed") | ||
| EOF |
Contributor
There was a problem hiding this comment.
This large Python script embedded using a heredoc is difficult to maintain, lint, and debug. For better code cleanliness and modularity, as per the repository's style guide, this logic should be moved into a separate Python script. The new script could then be called with the performance metrics and thresholds as command-line arguments.
References
- The style guide (lines 10-11, 61) emphasizes code cleanliness, modularity, and maintainability. Embedding a large script within another script goes against these principles, and the guide suggests that code that could be more modular should be fixed. (link)
Comment on lines
+102
to
+106
| if ! "${SCRIPT_DIR}/run-long-doc-qa-l2.sh"; then | ||
| echo "long doc QA L2 test failed" | ||
| TEST_RESULT=1 | ||
| exit 1 | ||
| fi |
Contributor
There was a problem hiding this comment.
The exit 1 on line 105 causes the entire test suite to terminate prematurely if this test fails. This is inconsistent with the error handling in other test steps in this script (e.g., for run-long-doc-qa.sh), which only set TEST_RESULT=1 and allow subsequent tests to run. To ensure all tests are executed and all failures are reported, the exit 1 should be removed.
Suggested change
| if ! "${SCRIPT_DIR}/run-long-doc-qa-l2.sh"; then | |
| echo "long doc QA L2 test failed" | |
| TEST_RESULT=1 | |
| exit 1 | |
| fi | |
| if ! "${SCRIPT_DIR}/run-long-doc-qa-l2.sh"; then | |
| echo "long doc QA L2 test failed" | |
| TEST_RESULT=1 | |
| fi |
Contributor
|
unblocked by: #2886 |
Contributor
|
jooho-XCENA
pushed a commit
to xcena-dev/LMCache
that referenced
this pull request
Apr 2, 2026
Signed-off-by: Oasis-Git <ayw.sirius19@gmail.com> Co-authored-by: Samuel Shen <slshen@tensormesh.ai>
jooho-XCENA
pushed a commit
to xcena-dev/LMCache
that referenced
this pull request
Apr 2, 2026
Signed-off-by: Oasis-Git <ayw.sirius19@gmail.com> Co-authored-by: Samuel Shen <slshen@tensormesh.ai>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What this PR does / why we need it:
Special notes for your reviewers:
If applicable: