Cache miss stampedes

[BUSTARRET, Jean-francois]

+1. We have the same problem.

We have many ways of accessing/sorting/filtering data (most viewed (1 day/1 week/1 month/...)/most commented/most recent/..., filtered by tags and/or group of users and/or ...).

Sometimes, the logic can be pretty complex, and I'm willing to cache (sometimes just to avoid "press F5" denial of services). But I can't deal with all de-caching patterns.

Having some kind of logic in the caching layer is way cheaper to implement, and solves most problems (and let's deal with the rest later, when we have time (which we won't) or when we'll be able to find someone competent enough to be hired (we won't, he'll go to Facebook)).

Jean-François

> -----Message d'origine-----
> De : memcached-bounces at lists.danga.com 
> 
> Well. Uh.
> 
> Sometimes ya just can't, yaknow? :)
> 
> In an ideal world your cache never expires, memcached never 
> flaps, and you have tools updating caches in the background 
> that work flawlessly. 
> Unfortunately developers don't always have the time to make 
> these perfect, but it's "easy" to patch in one of the 
> previous suggestions to deal with the problem.
> 
> 
> Steven Grimm wrote:
> > I admit I'm a bit baffled by this discussion (and I also 
> admit I have 
> > only been skimming it, so this might be a retread.) It 
> seems like one 
> > of two situations should be true: