[DUG] Why InterBase

[Paul Heinz]

Richard wrote:
> Application server layers are scalable, you can just add more of them.
> There are more and more people who wish to scale adding distributed
> caches as well (open source ones), where the entire database is sucked
> into memory and the distributed cache deals with everything, making it
> ever so much more scalable. They work, they are in operation, you just
> have to have a bit more money for machines that can have lots of memory.

Hrmm... how does that work? If you add lotsa application servers and they
all pull the database into memory, how to you keep these caches coherent
with changes in the database?

I would have thought that modern DBs engines, given enough RAM, will pull
all the hot working page set into cache anyway. That's what DB engine
developers seem to spend their day futzing trying to outsmart the OS. That's
going by reading the linux kernel mailing list anyway :-)

So I guess I'm left wondering what the real bottle neck the app servers are
attempting to mitigate is.

Is it disk to memory latency? Is it memory to network latency? Some other
latency? the laws of physics :-), etc...

TTFN,
  Paul.