You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
While running this exporter in one of my environments, I noticed that the initial spin-up time to do all of the ListMetrics can take some time depending on the namespace. The main reason in my case is that we have a large number of resources under a particular namespace (e.g. AWS/SQS).
To improve performance, I believe we can parallelize these calls by namespace. This should reduce start-up time.
To address API limits, some handling of throttling exceptions and retrying them would be in order. I imagine something along the lines of N goroutines (let's say, 10) are started to call ListMetrics, and configurable via the config file if the implementation wants to raise or lower it.
Thanks!
The text was updated successfully, but these errors were encountered:
While running this exporter in one of my environments, I noticed that the initial spin-up time to do all of the ListMetrics can take some time depending on the namespace. The main reason in my case is that we have a large number of resources under a particular namespace (e.g. AWS/SQS).
To improve performance, I believe we can parallelize these calls by namespace. This should reduce start-up time.
To address API limits, some handling of throttling exceptions and retrying them would be in order. I imagine something along the lines of N goroutines (let's say, 10) are started to call ListMetrics, and configurable via the config file if the implementation wants to raise or lower it.
Thanks!
The text was updated successfully, but these errors were encountered: