Two QuantLib routines made available in a grid computing environment

[Francesca Mariani]

Grid computing is becoming important in the financial industry. Symphony is a middleware environment to distribute loads on a computing grid. The QuantLib routines: Black and Scholes formula and Monte Carlo engine have been made available in the Symphony environment through the applications bs_symphony and mc_symphony that are available in the website: http://www.ceri.uniroma1.it/ceri/zirilli/w2. We ask the opinions of the blog users to the following question: there is a real interest in making available QuantLib or some other mathematical finance software in a grid computing environment?

[Luigi Ballabio]

Hi Francesca,

On Fri, 2008-02-29 at 06:30 -0800, Francesca Mariani wrote:
> The QuantLib routines: Black and Scholes formula and Monte Carlo
> engine have been made available in the Symphony environment through
> the applications bs_symphony and mc_symphony that are available in the
> website: http://www.ceri.uniroma1.it/ceri/zirilli/w2.

Interesting. Is the paper above only available as a web page, or do you
have a printable version? Also, do you have a bibliographic reference we
can add to the QuantLib site?

> We ask the opinions of the blog users to the following question: there
> is a real interest in making available QuantLib or some other
> mathematical finance software in a grid computing environment?

I guess there is. The question is how much the existing code should be
modified to fit in a grid environment, and how much this would affect
its use in a single-threaded program (I think the latter is still more
relevant, since for most purposes it is enough---and much easier---to
spawn multiple processes rather than threads.)

Another question: from the page you mention above, I seem to understand
that QuantLib was not modified at all---which makes me think that the
exported functionality was made available on the grid, but was not made
multi-threaded itself. Is this correct?

Luigi


--

Green's Law of Debate:
Anything is possible if you don't know what you're talking about.



-------------------------------------------------------------------------
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/
_______________________________________________
QuantLib-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-users

[Francesca Mariani]

Hi Luigi, below the answers to your questions: Question 1: The work described in http://www.ceri.uniroma1.it/ceri/zirilli/w2 is only available as web page there is not, at the moment, a printable version. However there is a different piece of work where a finance application (not belonging in QuantLib) has been made available in the Symphony environment where a printable paper exists [1]. Question 2: You are right the QuantLib subrourines that have been made available in the Symphony environment have not been modified. In particular they have not been made multi-threaded. Francesca [1] L. Fatone, F. Mariani, M.C. Recchioni, F. Zirilli: " Pricing realized variance options using integrated stochastic variance options in the Heston stochastic volatility model", Discrete and Continuous Dynamical Systems, Supplement 2007, (2007) 354-363 (http://www.econ.univpm.it/recchioni/finance/w4)

Luigi Ballabio wrote

Hi Francesca, On Fri, 2008-02-29 at 06:30 -0800, Francesca Mariani wrote: > The QuantLib routines: Black and Scholes formula and Monte Carlo > engine have been made available in the Symphony environment through > the applications bs_symphony and mc_symphony that are available in the > website: http://www.ceri.uniroma1.it/ceri/zirilli/w2. Interesting. Is the paper above only available as a web page, or do you have a printable version? Also, do you have a bibliographic reference we can add to the QuantLib site? > We ask the opinions of the blog users to the following question: there > is a real interest in making available QuantLib or some other > mathematical finance software in a grid computing environment? I guess there is. The question is how much the existing code should be modified to fit in a grid environment, and how much this would affect its use in a single-threaded program (I think the latter is still more relevant, since for most purposes it is enough---and much easier---to spawn multiple processes rather than threads.) Another question: from the page you mention above, I seem to understand that QuantLib was not modified at all---which makes me think that the exported functionality was made available on the grid, but was not made multi-threaded itself. Is this correct? Luigi -- Green's Law of Debate: Anything is possible if you don't know what you're talking about. ------------------------------------------------------------------------- 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/ _______________________________________________ QuantLib-users mailing list [e-mail] https://lists.sourceforge.net/lists/listinfo/quantlib-users

[Young Yoon]

Hi Luigi,

It is not necessary to modify QuantLib sub-routines. You can invoke the sub-routines in <a href="http://en.wikipedia.org/wiki/Symphony_Developer_Edition>Symphony service that can be deployed across many processors in SOA computational grid. Symphony client decomposes financial data and sends them over to the services in the grid for parallel computation.
As for the multi-threads on the service, you can use any programming tools, e.g., POSIX, openMP, to make your service multi-threaded and increase the performance gain even more. You can find more details about gridifying your financial application at Symphony Developer Edition user/developer community

- Young ([e-mail])