A view from Wipro NJ Office

Cooking Gas Subsidy in India

Why 350/-Rs subsidy for an owner of a BMW?

Oil marketing companies in India is loosing close to 350/-Rs per cylinder as subsidy !!! I don't understand why people like us require this 350/-Rs cut ? I think government should really think about differential pricing for this ...When affluent people can spend lot of money on partying and luxury items they will have 750/-bucks for their cooking gas also?

Grid Computing and Capital Markets

GRID COMPUTING: Fuel To the growing Derivative Industry
The cost of information is huge, when it comes to the financial industry every lag in computation can lead to enormous losses. One major value addition to the financial industry would be in the form of faster computation of complex and intensive financial models. With every passing day we can see more complex and sophisticated financial analytics models, most of these innovations hit the glass ceiling in terms of computation capabilities. The advent of Grid computation has been a great promising technology which can break the mismatch of our analytical model and computation capability.

What is Grid Computation?

Grid computation can be defined as distributed processing technique that spreads the computation over multiple hardware and software. The essence of grid computation is in collaboration; basically it breaks down the problem into number of smaller units and then distributes the units to different machines. The grid can be formed of disparate technology and platforms. Grid technologies are better used in a system where the process can be easily decomposed into sub processes and is not highly interdependent.
The best example of Grid computing is SETI@Home project. SETI@home is a scientific experiment that uses Internet-connected computers in the Search for Extraterrestrial Intelligence (SETI). People can participate by running free programs that download and analyze radio telescope data. At present more than 3 million users are hooked into this distributed computing effort.

The Grid technology coupled with the open source/standards has made inroads in various industry or vertical. Some of the most common application in the financial services industry can be elaborated as mentioned:

Derivative analysis:
The tremendous growth in derivative segment has created the need for high end financial data analytics for financial instruments such as Swaps, Options, Future, Exotic product etc. Derivative analysis is very critical and is an on going process for financial institutions to: calculate the pricing for the instruments, hedge the risk, identify any arbitrage opportunity, algorithmic trading etc. Some of the above analysis requires Monte Carlo simulation in Black Scholes formula. The analysis is based on floating point math’s operation such as logarithm, exponent, square-root, and division. In addition, these computations must be repeated over millions of iterations. The numerical Black-Scholes solution is typically used within a Monte Carlo simulation, where the value of a derivative is estimated by computing the expected value, or average, of the values from a large number different scenario, each represent a different market condition.
With grid computing in place many of these analytic and calculations have eased out and also helped the financial houses to work on thin margins.

Statistical computation:
The large scale data analysis and statistical calculation involved high speed program and hardware, altering these software is not advisable as there are issues related to the secrecy of data or compliance etc. The statistical analysis for financial instruments and complex product are notably stringent and need to be highly precise, involving intensive mathematical computation with millions of iterations and programmed in a high-level language that would require high cost in terms of altering. With distributed computing in place these statistical calculation can be eased.

Risk Analysis and calculation of VAR:
There has been intense pressure to create innovative product to increase the margins in derivative market. As a result, transactions continue to become more complex and more cumbersome to value. In terms of front office the requirements are limited to daily revaluation and hedge sensitivity calculations. In contrast, risk management analysis typically requires from hundreds to thousands of replications as the basis for estimating probability distributions.
If the use of front office pricing routines is imposed on risk systems, this seriously undermines the effectiveness of those systems when combined with realistic constraints on the computing power available for their operation. The essential key must be to apply analytical short-cuts, some of a very complex mathematical nature, to analyze well behaved transactions.

Scalable data warehousing for the large chunk of data used in analytics:
Create and maintain a flexible data infrastructure to respond to the constantly changing data requirements for modeling and reporting of risks and exposures.
The necessary data elements needed for effective risk calculations will evolve continuously both because of new businesses and new products and because of a desire to perform more complex calculations

With such diverse usage in the industry it has helped the trader’s as well financial institution to realize some common benefit in terms of:
Improvement in the time to market of information and its responsiveness
Provides more robust and resilient service as it spread application processing over many servers, if any sever goes down, the applications may run a little slower, but still completes successfully. Completing a job before the next working day is very important in financial world.
Improves the utilization of resource especially the idle resource

Conclusion:

Over the years, the grid technology has made significant inroads into the financial services industry. Currently companies in the financial sector depend heavily on such computationally intensive calculations to gain a competitive advantage. An improvement in VaR calculation time or high end statistical computation, be it front office or back office process, the traders and the financial institution with the implementation of grid computing has been able to reap more accurate results and better margin.