tokers commented on pull request #4965: URL: https://github.com/apache/apisix/pull/4965#issuecomment-911475113
> > > Could you give some use cases about this feature? > > > okay~ > > > In the simulation environment, we hope to use a small amount of real traffic to verify the new function, but the real traffic may be huge, which puts great pressure on the simulation service. > > Personally, I prefer to sample randomly than sample the first N. As, > > 1. sample randomly is more accurate. It doesn't miss requests after the first N > 2. we don't need a place to store the count. > > What about your opinion? > @membphis @tokers @tzssangglass I think a random sample is better than the current way. Also, if we support these rate-limiting ways, we have to change multiple plugins in the future if new rate-limiting algorithms are introduced, and it's prone to be inconsistent on the configurations items. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
