- From: Poul-Henning Kamp <phk@phk.freebsd.dk>
- Date: Mon, 02 Apr 2012 14:44:19 +0000
- To: Mike Belshe <mike@belshe.com>
- cc: Amos Jeffries <squid3@treenet.co.nz>, ietf-http-wg@w3.org
In message <CABaLYCttHxrCNsx6jCGG5e28yXfGX4K_GG-Mp6G-L7M8_p+tCg@mail.gmail.com> , Mike Belshe writes: >I agree we should pay attention to scalability - and we have. > >Please don't disregard that Google servers switched to SPDY with zero >additional hardware (the google servers are fully conformant http/1.1 >proxies with a lot more DoS logic than the average site). Google is not typical: A) Google delivers a very scalable service B) Google can and has scaled this out to datacenters throughout the globe Not all users of HTTP can do that. The benefits you cite, compression of headers and multiplexing of connections, can also be realized by other protocols, which might be more generally suitable than SPDY. But speaking of numbers: Have Google or anybody else published numbers for how fast they have managed to get SPDY to run on contemporary computer hardware ? How many requests per second ? How many concurrent connections ? How many DoS-attacks ? -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 phk@FreeBSD.ORG | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence.
Received on Monday, 2 April 2012 14:44:47 UTC