[dns-operations] Google Chrome Pre-Caching

Chris Roosenraad chris.roosenraad at twcable.com
Thu Sep 4 14:07:11 UTC 2008


On 9/3/08 8:22 PM, "Michael Sinatra" <michael at rancid.berkeley.edu> wrote:

> If your users tend to browse sites with lots of links, or if their
> default pages are have lots of links, your customer-facing DNS servers
> will get lots of DNS queries.  We were able to verify the behavior by
> having my boss load the default page for www.berkeley.edu.  His machine
> immediately queried for the hostnames for *all* of the links on that
> page, even though he visited not one of them.

These stats are by no means scientific, but I think give a nice baseline.

I did my "normal" surfing this morning twice, once with Firefox (with all
the various extensions turned off) and then again with Chrome (after
flushing the windows DNS cache).  While I'm not a typical home user, I think
the sample of websites is pretty good.  CNN, BBC, Washington Post, a few web
comics, a weather site, slashdot, the register.

Firefox generated 194 DNS packets (queries and responses)
Chrome generated 638 DNS packets (queries and responses)
Chrome was 3.289x the query load of Firefox.

Obviously that # will vary based on what sites you visit.  If you go to a
page with tons of links out (say a very well linked blog), you could easily
have a 100-1 ratio.  Go to an internal facing one, and I bet you'll be
closer to 1.5-1 or maybe lower.

Still, if you assume that between 1/2 and 2/3 of all DNS queries come from
browser activity...and if you say that Chrome will result in between 3 and 4
times the DNS queries...then you can expect for every 1% of your customers
who switch to Chrome, your total DNS traffic will increase 1% to 2%.

Now, there are lots of assumptions in that math...but I think its a good
starting point.

Now, as David Dagon pointed out, there is an estimated 2% take rate so far.
I doubt it'll climb much higher for a while...the early adopters have
switched, and the rest of the folks out there have something they like...so
why change?  If its really successful, I'd say maybe 5% by EOY.

So...with all that, I agree with David.  Not something to serious worry
about right now...but an interesting "change" in customer behavior.  The
only folks I think have to worry are those who run the .co TLD...they're
going to get hammered by this.

Chris R. Roosenraad
Principal Engineer
Time Warner Cable - ATG
13241 Woodland Park Road
Herndon, VA  20171
+1 (703) 345 3438
chris.roosenraad at twcable.com

This E-mail and any of its attachments may contain Time Warner
Cable proprietary information, which is privileged, confidential,
or subject to copyright belonging to Time Warner Cable. This E-mail
is intended solely for the use of the individual or entity to which
it is addressed. If you are not the intended recipient of this
E-mail, you are hereby notified that any dissemination,
distribution, copying, or action taken in relation to the contents
of and attachments to this E-mail is strictly prohibited and may be
unlawful. If you have received this E-mail in error, please notify
the sender immediately and permanently delete the original and any
copy of this E-mail and any printout.

More information about the dns-operations mailing list