mirror of
https://github.com/BerriAI/litellm.git
synced 2026-09-08 22:21:35 +00:00
reduce max memory tests from 100k/250k to 50k for faster CI
- Rename test_memory_baseline_100k to test_memory_baseline_50k (router) - Rename test_proxy_memory_baseline_250k to test_proxy_memory_baseline_50k (proxy) - Update CI configuration to use new 50k test names - Reduces test time while still providing comprehensive scale testing - Router 50k test: ~90 seconds vs 3-4 minutes for 100k - Proxy 50k test: ~2.5 minutes vs 16 minutes for 250k
This commit is contained in:
parent
3fe36cee72
commit
1cce3822c5
3 changed files with 12 additions and 12 deletions
|
|
@ -2069,8 +2069,8 @@ jobs:
|
|||
python -m pytest tests/load_tests/test_router_acompletion_memory_growth.py::test_memory_baseline_10k -v -s --tb=short
|
||||
echo "Running test_memory_baseline_30k..."
|
||||
python -m pytest tests/load_tests/test_router_acompletion_memory_growth.py::test_memory_baseline_30k -v -s --tb=short
|
||||
echo "Running test_memory_baseline_100k..."
|
||||
python -m pytest tests/load_tests/test_router_acompletion_memory_growth.py::test_memory_baseline_100k -v -s --tb=short
|
||||
echo "Running test_memory_baseline_50k..."
|
||||
python -m pytest tests/load_tests/test_router_acompletion_memory_growth.py::test_memory_baseline_50k -v -s --tb=short
|
||||
no_output_timeout: 30m
|
||||
- run:
|
||||
name: Run Proxy Server Memory Growth Tests
|
||||
|
|
@ -2084,8 +2084,8 @@ jobs:
|
|||
python -m pytest tests/load_tests/test_proxy_chat_completions_memory_growth.py::test_proxy_memory_baseline_20k -v -s --tb=short
|
||||
echo "Running test_proxy_memory_baseline_30k..."
|
||||
python -m pytest tests/load_tests/test_proxy_chat_completions_memory_growth.py::test_proxy_memory_baseline_30k -v -s --tb=short
|
||||
echo "Running test_proxy_memory_baseline_250k..."
|
||||
python -m pytest tests/load_tests/test_proxy_chat_completions_memory_growth.py::test_proxy_memory_baseline_250k -v -s --tb=short
|
||||
echo "Running test_proxy_memory_baseline_50k..."
|
||||
python -m pytest tests/load_tests/test_proxy_chat_completions_memory_growth.py::test_proxy_memory_baseline_50k -v -s --tb=short
|
||||
no_output_timeout: 45m
|
||||
|
||||
db_migration_disable_update_check:
|
||||
|
|
|
|||
|
|
@ -426,9 +426,9 @@ async def test_proxy_memory_baseline_30k(proxy_server, limit_memory):
|
|||
@pytest.mark.asyncio
|
||||
@pytest.mark.limit_leaks(MEMORY_LIMIT)
|
||||
@pytest.mark.no_parallel # Must run sequentially - measures process memory
|
||||
async def test_proxy_memory_baseline_250k(proxy_server, limit_memory):
|
||||
async def test_proxy_memory_baseline_50k(proxy_server, limit_memory):
|
||||
"""
|
||||
Memory baseline test with 250,000 requests to the proxy server.
|
||||
Memory baseline test with 50,000 requests to the proxy server.
|
||||
Uses @pytest.mark.limit_leaks("40 MB") to enforce memory limit.
|
||||
If test_proxy_memory_baseline_1k and test_proxy_memory_baseline_2k pass but this fails,
|
||||
it's a clear sign of sequential/progressive memory growth.
|
||||
|
|
@ -436,6 +436,6 @@ async def test_proxy_memory_baseline_250k(proxy_server, limit_memory):
|
|||
NOTE: This test should be run INDIVIDUALLY, not with other tests in this file.
|
||||
Running multiple tests together causes memory baseline drift, making it difficult
|
||||
to accurately detect linear memory growth. Run with:
|
||||
pytest tests/load_tests/test_proxy_chat_completions_memory_growth.py::test_proxy_memory_baseline_250k -v
|
||||
pytest tests/load_tests/test_proxy_chat_completions_memory_growth.py::test_proxy_memory_baseline_50k -v
|
||||
"""
|
||||
await run_proxy_memory_baseline_test(250000, proxy_server, limit_memory)
|
||||
await run_proxy_memory_baseline_test(50000, proxy_server, limit_memory)
|
||||
|
|
|
|||
|
|
@ -125,9 +125,9 @@ async def test_memory_baseline_30k(test_router, limit_memory):
|
|||
@pytest.mark.asyncio
|
||||
@pytest.mark.limit_leaks(MEMORY_LIMIT)
|
||||
@pytest.mark.no_parallel # Must run sequentially - measures process memory
|
||||
async def test_memory_baseline_100k(test_router, limit_memory):
|
||||
async def test_memory_baseline_50k(test_router, limit_memory):
|
||||
"""
|
||||
Memory baseline test with 100,000 requests.
|
||||
Memory baseline test with 50,000 requests.
|
||||
Uses @pytest.mark.limit_leaks("40 MB") to enforce memory limit.
|
||||
If test_memory_baseline_1k and test_memory_baseline_2k pass but this fails,
|
||||
it's a clear sign of sequential/progressive memory growth.
|
||||
|
|
@ -135,6 +135,6 @@ async def test_memory_baseline_100k(test_router, limit_memory):
|
|||
NOTE: This test should be run INDIVIDUALLY, not with other tests in this file.
|
||||
Running multiple tests together causes memory baseline drift, making it difficult
|
||||
to accurately detect linear memory growth. Run with:
|
||||
pytest tests/load_tests/test_router_acompletion_memory_growth.py::test_memory_baseline_100k -v
|
||||
pytest tests/load_tests/test_router_acompletion_memory_growth.py::test_memory_baseline_50k -v
|
||||
"""
|
||||
await run_memory_baseline_test(100000, test_router, limit_memory)
|
||||
await run_memory_baseline_test(50000, test_router, limit_memory)
|
||||
|
|
|
|||
Loading…
Add table
Reference in a new issue