Re: Thesis about OpenSSI

Hi...

On Mon, May 26, 2008 at 7:34 AM, Fabian Herb
<[email protected]> wrote:
> Dear users and developers of OpenSSI,
>
> my name is Fabian Herb and I study at the RWTH Aachen University. I am
> working on a thesis about history and future of OpenSSI and SSI
> clustering in general. I really would appreciate if somebody could
> help me by answering one or two of these questions:
>
> 1. Where do you use OpenSSI? I'd like to know wether it is in
> commercial or academic use, in testing or production environment, and
> which kind of applications you run on your cluster, and of course how
> many and which types of machines it consists of.

Once at my home lab. It was purely for experiment purposes. I ran
application such as povray, self made Monte carlo application.

It was just 2 nodes cluster. And lately I convert it into 3 virtual
nodes, that is running inside Qemu. It was slow, but enough for
testing purposes.


> 2. What do you think does the future of OpenSSI look like, now that
> support from HP has ceased?

I think it has a bright future, because even multi core processors get
cheaper, eventually people will need very fast computation...faster
than current standart. And that's only feasible and economically
doable by constructing a computer cluster. This is where openSSI plays
the role, to make cluster building easier and easy to manage (by
gathering them into single system image).

Even HP has no longer support it, it will still live, at least for the
code base. It depends on how the open source community takes this code
and develop it further. Certainly not easy, since kernel level
clustering like openSSI requires somebody with all round kernel
understanding.

regards,

Mulyadi.

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/