Hi,<div><br></div><div>I was looking into varnishtop to find performance bottlenecks on a website. Thinking about it I think a useful feature is currently missing from this tool. (while it _has_ all the available raw data).</div>
<div>I'd like to discuss it here, to get your opinion.</div><div><br></div><div>Currently varnishtop shows the top urls purely based on frequency. Depending on your usecase that might be usefull, but might also NOT give you the bottlenecks you are looking for.</div>
<div><br></div><div><br></div><div>I'd like to use varnish for the following (example) scenario:</div><div><br></div><div>Consider an website site with many dynamic urls. (dynamic means varnish does not cache them, they are always passed through to the backend)</div>
<div>Of those there are two urls that are behaving "badly" X and Y (that is they are slow: lets say 400ms)</div><div>We don't know that _these_ to are behaving badly.</div><div><br></div><div>X: is called very low frequently -> so this is NOT a perf bottleneck.</div>
<div>Y: is called often -> THIS IS THE BOTTLENECK -> this is the url that brings down the site.</div><div>Z: is called often -> this one is fast -> not a bottleneck.</div><div><br></div><div>I'd like varnishtop to show Y, (and not X-like urls( urls that are not called often), and not Z-like urls (urls that are fast).</div>
<div><br></div><div>For this to work varnishtop should (optionally ofcourse) order the url not by request/rate, but rather by "total time spend" on this url. Something like the product of freq and avg time on an url.</div>
<div><br></div><div>in this case</div><div>expensive url != often called backend url</div><div>expensive url != slow backend url</div><div>expensive url = backend url with high total time spend (freq x duration) (So the slow ones, that are called often)</div>
<div><br></div><div><br></div><div>What do you guys think ?</div><div>This would instantly give the problem areas of a website, no ?</div><div><br></div><div><br></div><div>Regards,</div><div>Harm Verhagen</div>