[Beowulf] SiCortex experience anyone?

[Perry E. Metzger]

Joe Landman <landman at scalableinformatics.com> writes:
>> I think you're looking through the lens of your customer base. A large
>
> Quite possibly.
>
>> number of clusters never run commercial codes, and some commercial
>
> True.  They tend to be low margin commodity clusters that are very
> hard to support "value added" companies atop though.

That depends. SiCortex claims that they're an overall win for their
customers. Perhaps this is true, perhaps not, but one can certainly
see how people who need extremely large clusters might care a lot.

>> codes come with source. So the addressable market is still pretty big.
>
> Well... the TAM is "large" in some sense, but the margins in the
> academic/research market are nothing to write home about.  It is hard
> to build a (venture backed) company servicing those markets.

There are a lot of venture backed companies in computational chemistry
that seem to make all their money on academic and research
-- indeed, I don't know of any market for computational chemistry
software outside of academia and research (though of course some of
that research is paid for by pharma companies). Going outside of
computers, there are plenty of companies that make their money selling
biotechnology equipment and specialty chemicals that are primarily of
interest in academia and in research -- arrays to measure gene
expression levels, for example, but there is a very broad market.

None of this should be surprising -- there is a lot of money spent on
academic and commercial research in the world.


Perry
-- 
Perry E. Metzger		perry at piermont.com