On Mon, Aug 13, 2012 at 8:06 AM, Johan Euphrosine <[email protected]> wrote: > >> I tried both. Going ab -> frontend -> backend is _much_ slower than >> going directly to a backend. > > And the numbers you posted were for hitting directly the vps instance? or > urlfetch from a frontend?
I'm not sure which exact numbers you are referring to. The VPS tests run 'ab' on the same VPS running the server. The 'wh-test' noop tests ran 'ab' across a 30Mbit link with fairly low latency to ghs.google.com... basically the best connection I had available. Given that these are high-concurrency tests, I wouldn't expect latency to be a significant issue - there's a large pending backlog on the server in each case. Jeff (sorry, I've been travelling in internet dark zones for the last several days) -- You received this message because you are subscribed to the Google Groups "Google App Engine" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/google-appengine?hl=en.
