[Beowulf] (no subject)

[apletche at cisco.com]

So to echo Mark, one of the more prevalent technologies being bantered
about is the whole virtualization capabilities to server farms.  From a
clustering perspective this applies to the who utility computing
clusters where decisions are made by a virtualization manager that
handles hardware profiles, application profiles that identify hardware
profile requirements and the like...

Drew



> > grids are based on the absolutely mistaken premise that computing is

> > commodified and generic. you just arrange the plumbing and the 
> > flops will flow to whereever they're needed. I've never quite been 
> > clear on whether gridophiles believe this, and understand that there

> > are different instruction sets, different cache sizes, different 
> > clocks, different sizes and speeds of memory, etc. maybe they're 
> > simply comfortable working with the least common denominator - a 
> > portable language like java/perl/etc and nothing but embarassingly
parallel codes.

> >
> > A nice use case is to use grid stuff to get a uniform way to access
preinstalled applications,
> > locally tuned according to the idiosyncrasies of the local systems.

> oh, absolutely. in this case, a "grid" is just an application farm.

> and I'm sure there's lots of demand for that. but it assumes that your
apps change very slowly or you > have a _horde_ of very effective admins
who can make all the grid nodes look basically identical to
> the app.

> I guess I should have mentioned that my own context is
academic/research HPC, where nearly every job > runs a unique
executable, and for the most part, the only "applications" installed are
compilers ;)

> that's not entirely true - we have some users who run Gaussian, but
they tend to be fairly limited in > number. our focus is actually to get
researchers to think bigger, which usually means they can't use > some
off-the-shelf app.

> in that way of thinking, grids make a lot of sense as a
shrink-wrap-app farm.

> regards, mark hahn.