[PEAK] The path of model and storage

[Robert Brewer]

John D. Heintz wrote:
> Robert: are you asking about the equivalent of a read-through 
> cache for a database?

Basically, yes.

> Hibernate (an object-relational bridge layer) supports two levels of 
> caching: Session and clustered. See http://hibernate.org/21.html#A43

Looks interesting (for the next time I want to feed my masochism with
Java ;).

> I suspect that Robert is interested in the same thing I am:
> What is the PEAK-ish way of layering data caching into a process and 
> across processes?

Yup, although I'm only interested in the PEAK-ish way academically--I've
already written my own framework. It doesn't do cross-process any
differently than PEAK does, but it allows deployers to decide whether to
persist domain objects per-thread (= per-HHTP-request for most of my
apps), per-session, or per-app, either sharing them between users or
not.


Robert Brewer
MIS
Amor Ministries
fumanchu at amor.org