Gervase Markham wrote: > Brendan Eich wrote: > >>> Assume [...] that each Mozilla user's browser checks for the icon >>> once a week - say once every 100 page loads. >> >> Why are you assuming any such thing? Evidence? > > A Scientific Wild-Ass Guess (SWAG) based on what I know about how the > favicon system caches misses, the fact that most users close their > browser when they are not using it,
Here you were assuming that the not-found cache is lost at the end of a session. That's not the case. The XBL involved uses the disk cache, so the only ways a not-found entry is lost are through cache-clearing by the user, and through eviction due to the replacement policy, when the entry becomes older than any other and no free space is left in the disk cache quota. The disk cache quota is 50MB by default, so you'd have to spend 50MB of bandwidth on pages other than those for a site lacking a favicon to lose that site favicon's not-found entry. Or so the code intends -- there may be bugs still. These facts suggest that the bandwidth and log space consumed by unsolicited favicon server hits that result in 404 not found may be insignificant. We need to measure what actually happens with the pref enabled for 0.9.7. > and what seemed like sensible website usage per user for a random > high-traffic site (100 page views per session.) I have no idea if that's representative, but do you mean 100 distinct sites visited per session (page views != "site views", and the favicon probe will happen at most once per site, per session under any caching scenario)? Anyway, the not-found cache persists across sessions. > It was for illustrative purposes only - are my numbers more than an > order of magnitude off? Yes, if the cache works differently from how you assume -- you had 300MB (b for byte in your message, I'm assuming) of log space wasted on a month's worth of favicons not found, for a site that gets 15M hits/month, at 400 bytes/log-message. That's 300MB/400B or 750K messages, but 750K is 1/20th of 15M -- so you seem to be assuming every 20th hit gets a favicon not found. What's behind that assumption? 50MB of disk cache is a lot. It may be that favicon-not-found entries expire so rarely that the hit rate and consequent log space for any server is tiny. We should measure, again. Modeling the costs seems hard to me right now, and not as fruitful as actual measurement. /be > > > Gerv >
