Since what I care is latency, protocol overhead is insignificant in my case. It's odd, but is the best measure so far.
As ghs.* gets resolved to the nearest DC (which in my case is EU one) requests are "piped" throught it to App Engine servers. This EU-DC > US-DC > EU-DC thing is what makes apps slow to European users. As an illustration: $ ping ghs.google.com PING ghs.l.google.com (173.194.71.121) 56(84) bytes of data. 64 bytes from lb-in-f121.1e100.net (173.194.71.121): icmp_req=1 ttl=44 time=90.2 ms $ time wget http://$GAEAPP/ ... real 0m0.756s user 0m0.008s sys 0m0.004s $ time wget http://$GAEAPP/ ... real 0m0.198s user 0m0.000s sys 0m0.004s In last "wget" page got egde-cached (served from EU front-end) while 250-800+ ms is an "actual" latency to the app (paid, zero traffic, resident instance). P.S: Have you tried to post your question on StackOverflow? On Tue, Jan 29, 2013 at 2:33 AM, alex <[email protected]> wrote: > > Hey Igor, > > I'm not sure what you mean: ping and wget really use different > protocols so it's probably not a fair comparison. Plus, AFAIK > ghs.google.com and ghs.googlehosted.com resolve to different IPs based > on your location (correct me if I'm wrong), so this too wouldn't be > fair. And another thing is, myapp.appspot.com is usually a CNAME to > ghs.google.com, googlehosted.com or similar (at least, in my setup), > so myapp.appspot.com and ghs.google.com is the same thing. > > On Mon, Jan 14, 2013 at 11:07 PM, Igor Kharin <[email protected]> wrote: > > Oh, lucky you! Does it feels, well, less crowded there? :) > > > > And more importantly, what are latencies? Like, approximately, what's > > the delta between "ping ghs.google.com" and "time wget http://$EUAPP";? > > From what I see with edge-cached pages it might be around 30+ ms. > > > > Thanks. > > > > On Mon, Jan 14, 2013 at 9:12 PM, alex <[email protected]> wrote: > >> I've been playing around with an app hosted on EU datacenters for the > >> past few days and I've gotta say it's been a freaking awesome > >> experience so far, especially for what concerns latency. > >> > >> It's been so great that I think we'll go live next month with this app. > >> There's one thing I'm not sure of: it's a TT program as far I as can > >> tell so, when there's an issue does one create it as usually on the > >> issue tracker, even if it's a production one? > >> > >> For instance, there's something weird going on with the memcache > >> service right now, it shows all 0 and says "Could not retrieve > >> statistics. Try again shortly." > >> > >> thanks. > >> > >> -- > >> 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. > >> > > > > -- > > 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. > > > > -- > 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]. > Visit this group at http://groups.google.com/group/google-appengine?hl=en. > For more options, visit https://groups.google.com/groups/opt_out. > > -- 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]. Visit this group at http://groups.google.com/group/google-appengine?hl=en. For more options, visit https://groups.google.com/groups/opt_out.
