Saturday, December 1, 2007

One milisecond faster - IBM's Stream Computing

IBM is developing a project called System S, which promises wonders and, among other things, will revolutionize the speed of processing data so as to project computing systems into a new era. 

The letter "S" comes from the word streaming. The project was announced in a yearly technology fair in New York this month which was organized by SIFMA (Securites Industry and Financial Markets Association).

The news spread worldwide quickly because the most impressive feature about this system is that it is capable of analysing huge amounts of data in real time, including video streams.

One of the most important discoveries has to do with finding a way to send different work loads to the most appropriate computing hardware for this job, such as the case of a broad band cell processor capable of managing video streams.



Having the option of using and being able to compile an enormously greater quantity of data, now it is possible to correlate information, which before it was practically unfeasable to do in a more precise way due to the fact that each computer system was different and could not integrate with the other since the data were first stored and then available for accessing, retrieval and reporting.


System S is designed especially to support the financial services industry because in it one milisecond counts greatly. This means not only that in such a fraction of a second you could loose millions in terms of money, but also that all the market opportunities that were available at that moment, would have been lost.


Current systems act mainly as independent islands and are not part of a bigger whole, therefore, they cannot be used effectively to make realistic forecasts in time.



This new system offers the posibility of thinking and acting in short or very short term events so that projects can be planned and executed way before the competition, which, for now, is looking at data from a previous time period in order to make strategic decisions and only then being able to execute them.


The next step is to work directly with clients to build applications that will make up the foundation of a new and powerful method that allows streams, waves, or, in this case, tsunamis loaded with data and turn them into hugely useful information for fields such as finance, medicine, astronomy, and security. 

Concept Copyright Esteban Dobronsky 2007 ©


Click on link for all posts from latest entry