isapir commented on PR #760: URL: https://github.com/apache/tomcat/pull/760#issuecomment-2403736338
> In Performance Test, request client threads have sleep ops. Why? Can those be removed? > Average cost of each rateLimiter is about 3500ns. The difference in time cost between Fast and Exact rate limiters is less than 0.001 milliseconds How do you measure exactly? > Shall we make ExactRateLimiter as a default implementation? I still prefer multiple implementation. For me the most important thing is DDoS mitigation, and under an attack every little bit helps. > According to the performance test results, for FastRateLimiter, the configuration of 120/4s is actually converted to 123/4s. Then the number of requests passing through each IP is more than that of ExactRateLimiter, and the number of blocked requests is less than that of ExactRateLimiter, and the throughput will decrease. Can you test with a much higher value so that the number of requests allowed is not reached? -- 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: dev-unsubscr...@tomcat.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
