autotune
autotune copied to clipboard
Autonomous Performance Tuning for Kubernetes!
Error handling needs to be uniform across all failure conditions. At present, there are exception stack traces logged for some of the cases.
I see the below exception in kruize pod log, can we handle this or move it under debug info. ``` Exception in thread "OkHttp Dispatcher" java.util.concurrent.RejectedExecutionException: Task java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask@a0fa2ea[Not completed, task...
- Create Experiments doesn’t throw an error when conflicting fields are present in the json posted to create the experiment. Conflicting fields: - performanceProfile and slo - Selector and deployment...
Currently list recommendations API doesn't return the recommendations over a range of time period (start to end). The user need to manually search for them in the recommendations list
Recommendation duration "terms" to be part of the Kruize object and configurable through API
Explore time series [forecasting](https://www.timescale.com/blog/what-is-time-series-forecasting)
Usage of memRSS for Memory
Fixing percentile dynamically based on usage+throttle percentile distribution Epic #716 5.i
There are few datapoints for a pod from the cluster when there is no cpu and memory usage. Do we need to consider this timestamp in the monitoring duration for...