sshivaditya2019 / test-public

0 stars 0 forks source link

Review Cache Strategies for Java Service Layer to Handle High Load #111

Open sshivaditya2019 opened 1 month ago

sshivaditya2019 commented 1 month ago

Description: The Java service layer's performance deteriorates under high load, partly due to the absence of effective caching. Evaluate and implement advanced caching strategies to ensure the service layer performs optimally during peak usage. Steps to Reproduce: Simulate high load on the Java service layer. Analyze performance metrics and identify bottlenecks. Expected Behavior: Service layer maintains stability and performance under high load with improved caching strategies. Current Behavior: Performance degradation and potential instability under high load. Attachments: N/A

test-app-ubo[bot] commented 1 month ago

[!NOTE] The following contributors may be suitable for this task:

sshivaditya

78% Match sshivaditya2019/test-public#112

sshivaditya2019

78% Match sshivaditya2019/test-public#112

0x4007 commented 1 month ago

Please test with the same exact task specifications that we had problems with. It lets us compare apples-to-apples.

sshivaditya2019 commented 1 month ago

Please test with the same exact task specifications that we had problems with. It lets us compare apples-to-apples.

For issue deduplication, I’ve reviewed the existing issues related to the stacking problem. However, I’m unable to find any issues that relate to the same problem with task-matching. Could you please direct me to those issues?

0x4007 commented 1 month ago

However, I’m unable to find any issues that relate to the same problem with task-matching. Could you please direct me to those issues?

I don't understand this

sshivaditya2019 commented 1 month ago

However, I’m unable to find any issues that relate to the same problem with task-matching. Could you please direct me to those issues?

I don't understand this

I’m having trouble finding instances where task matching didn’t work as expected. I tested some cases where deduplication was problematic on the updated model and shared the results through the QA link in the PR.