Supercomputing in Nepal

Place where young entrepreneurs can come together, share information, network, seek new business and ideas
kazi
Posts: 978
Joined: Mon Jun 29, 2009 11:55 am

Supercomputing in Nepal

Postby kazi » Mon Jul 06, 2009 10:04 am

Assuming that electricity is not a problem, what is the cheapest configuration to run a supercomputing (quasi-supercomputing) service?
--
"Mother and motherland are more precious than heaven." But that does not mean we must cling to our mothers. The least I can do for Nepal is to bring awareness among the Nepali people. And this Nepali forum is the platform for me.

Ganesha
Posts: 634
Joined: Mon Jun 15, 2009 8:51 am
Contact:

Re: Supercomputing in Nepal

Postby Ganesha » Fri Sep 04, 2009 12:30 am

First, you will need to explain what "quasi - computing" is all about. A simple google search is not giving me a lot to read. Anybody helping? :)
aawartan.org - nepali forum

kazi
Posts: 978
Joined: Mon Jun 29, 2009 11:55 am

Re: Supercomputing in Nepal

Postby kazi » Fri Sep 04, 2009 9:46 pm

Well, even as laymen we do not need to define super-computing. With this assumption we can simply add quasi- to refer to a computing service that offers comparable capacity (speed and quantity), but perhaps without dedicated hardware and high costs. For example, using the cloud, virtualization, grid, etc.
--
"Mother and motherland are more precious than heaven." But that does not mean we must cling to our mothers. The least I can do for Nepal is to bring awareness among the Nepali people. And this Nepali forum is the platform for me.

Ganesha
Posts: 634
Joined: Mon Jun 15, 2009 8:51 am
Contact:

Re: Supercomputing in Nepal

Postby Ganesha » Sat Sep 05, 2009 10:11 pm

In that case, we need to find out how Nepal has a competitive advantage in bringing up such a service. If we can build such machine cheap in Nepal, perhaps there is a market. The government might want to lease it as it is digitizing Land Revenue Offices, issuing Machine Readable Passports, creating a new system for vehicle number plates etc. (These processes might not be compute intensive today but they show a definite path of the government that it is aware of IT and will need much more processing power in the days to come). Though we might use our brainpower to design it and use our manpower to assemble it, we will still have to buy the hardware (machine and network infrastructure) which happens to be the most expensive part of the equation. I don't know how cheap such computing is as provided today by the likes of Amazon and Synaptic Hosting and Enomaly and others but the Nepali service should be made cheaper than them to begin with.
aawartan.org - nepali forum

kazi
Posts: 978
Joined: Mon Jun 29, 2009 11:55 am

Re: Supercomputing in Nepal

Postby kazi » Tue Sep 08, 2009 8:26 am

Yes, cost is always a driving force.. almost always. In cases of government usage, or scientific research, I think an in-house system is more favorable than external solutions. Two things come to my mind:

1. Of course the major objective of such a venture would be to have some return on investment. Eventually the venture should yeild some benefit. And in case of a successful venture, benefit is undoubted. Its a sure shot provided the project succeeds because when the power of computing is available right at your doorstep, the need for computing will be more urgent. Just as the baby cries more when (s)he sees the milk bottle, the online community will find ways to use the computing services. Just for the humor of it, why is China and Japan leading their own missions to the moon? Why dont they use the technologies that the US, India and Russia have? Why does Europe have to develop their own accelerator instead of just using the one at Fermi lab? Its not about cost only.

2. I do not think cost of hardware is really high in this day and age. If we really think about it, its the running and maintenance costs that shadow any other costs. So, if we can find a way to generate the running costs in the long run, the initial setup and hardware should not be unreasonable.. maybe for a single person its too much, but with grants, donations and investments.. Maybe its just optimistic speculation, but there'll be a start when we have a reasonable estimate. Besides monetary valuation, the bigger cost in such a venture would be intellectual. Will people be ready to contribute their brain power?
--
"Mother and motherland are more precious than heaven." But that does not mean we must cling to our mothers. The least I can do for Nepal is to bring awareness among the Nepali people. And this Nepali forum is the platform for me.


Return to “The Young Entrepreneur”

Who is online

Users browsing this forum: Google [Bot] and 1 guest