matrixorigin
[Bug]: There was a big difference about tpcc 100-1000 between running alone and running daily regression.
Is there an existing issue for the same bug?
- [X] I have checked the existing issues.
Branch Name
main
Commit ID
6ff08d6cff30aa00869325504631f2d5d6ef1ae0
Other Environment Information
- Hardware parameters:
- OS type:
- Others:
Actual Behavior
tpcc 100 1000位于daily regression流程中： job url： https://github.com/matrixorigin/mo-nightly-regression/actions/runs/8709270996/job/23892311885
profile：
2024-04-16_19_13_43.zip
单tpcc 100 1000流程： job url: https://github.com/matrixorigin/mo-nightly-regression/actions/runs/8716115767/job/23909381912
profile: 2024-04-17_12_34_18.zip
Expected Behavior
No response
Steps to Reproduce
trigger daily regression and binary search regression on tke.
Additional information
No response
The same resources running mixed complexity will definitely perform better than running a single load. Our memory and disk cache, then, pollute each other. This is not a bug.
0508 测试结果:
commit id：6e94513897f3a739602429e5a61d7c9127fd9a2a
单独跑tpcc 500-1000:
job URL：https://github.com/matrixorigin/mo-nightly-regression/actions/runs/8995415575/job/24710552822
混合跑tpcc 500-1000:
job url：https://github.com/matrixorigin/mo-nightly-regression/actions/runs/8951239289/job/24587949571
单独跑：
job url：https://github.com/matrixorigin/mo-nightly-regression/actions/runs/9026433240/job/24804803968
混合跑：
job url：https://github.com/matrixorigin/mo-nightly-regression/actions/runs/9019459807/job/24803217074
profile太大了，单独联系我
