Cache Invalidation and Database dependency?


Mandeep Singh just asked about cache invalidation policies in feedback on a previous entry.


Of course there is no universal answer here, it will totally depend on the specifics of your application, but I’d still like to tackle that question in greater length at some point. For now though, I wanted to quickly give some pointers to what I’ve already written on this.  Here is some decent guidance on caching in the PAG, see Chapter 3, under “Caching” and again in Chapter 4.


I could also do a general discussion in a blog to walk through some choices though to augment this information somewhat.


If there is feedback as to what direction you might like an expanded article to take feel free to put that here.  There’s plenty of meat on this topic.

Comments (1)

  1. Mandeep says:

    Thanks Rico,

    I went through those documents and those documents are a good begining . Also Caching Architecture Guide lists pros and cons of each.

    I am proposing a cheap mechanism (in my mind) of implementing WHIDBEY kind of cache( trigger and table based) and then for every object that is being cached witing a simple proc on lines (if exists (select * from changedtable where tableName=’x’) return true.

    If can blog on these lines or general discussion with which we know costs vs benifits of taking different approaches , that would be really helpful. I think this is one of the many important points to be considered seriously in order for project to be successful.

    Regards,

    Mandeep

Skip to main content