Ben Strong has done an analysis of how Google provides short load times. He says that Google and MS are violating RFC-3390. Readers of his blog have added extensive comments on why this may be occurring and if it is good or bad.
Ben Strong's blog entry "Google and Microsoft Cheat on Slow-Start. Should You?":
http://blog.benstrong.com/2010/11/google-and-microsoft-cheat-on-slow.html
mnot's blog entry "Digging Deeper with htracr":
http://www.mnot.net/blog/2010/11/27/htracr
Google has a technical paper addressing this titled "An Argument for Increasing TCP's Initial Congestion Window".
http://www.google.com/research/pubs/pub36640.html
My subjective impression is that Google page loads have become sluggish over the past year.
|
George R wrote:
> My subjective impression is that Google page loads
> have become sluggish over the past year.
I'm surprised to hear that. For many years, Google page loads have been pretty good. But over the past year, for me, they have become lightning fast. I'm in the UK.
That excludes the ultra-heavy pages from Knol, which have always been slow. |
