[erlang-questions] Amazon S3: Now with locking, transactions and caching

[Christopher Baus]

> I also coupled the virtual S3 table with a fixed-size cache that is
> built on top of a regular Mnesia table. All writes go trough the
> cache, which ensures that hot data is available immediately. So long
> as the cache API is used, any cache misses are automatically
> redirected to S3.

So all writes are written directly to S3?  Are you putting each write into
a separate object?  I've been doing some research into using S3 as a
general purpose data store which lead me to mnesia and erlang.  It seems
to me that the limiting factor is the transactional costs per write.

I thought not being able to do record appends was a real deal killer for
more structured data stores, but it sounds like you've figured out some
way around this.

I'd love to hear more about this.

Chris