JP Morgan supercomputer offers risk analysis in near real-time

[Andy Nguyen]

JP Morgan is now able to run risk analysis and price its global credit portfolio in near real-time after implementing application-led, High Performance Computing (HPC) capabilities developed by Maxeler Technologies.

JP Morgan uses mainly C++ for its pure analytical models and Python programming for the facilitation. For the new Maxeler system, it flattened the C++ code down to a Java code. The company also supports Excel and all different versions of Linux.

Back in 2008, it took eight hours to run the end-of-day risk process, a significant part of which was around moving data, manual data clean-up and verification.

The risk calculation time has now been reduced to about 238 seconds, with an FPGA time of 12 seconds.

http://www.computerworlduk.com/news...puter-offers-risk-analysis-in-near-real-time/

 

[Vuong Hoang Kim]

Thanks for sharing! More motivation to learn programming language ^^.

 

[ExSan]

no wonder why C is the most important language, though Cuda should be seriously considered from now on
12 / 238 = 0.05 ---- 5%

 

[Daniel Duffy]

no wonder why C is the most important language, though Cuda should be seriously considered from now on

C is h/w independent, CUDA not.
Just a remark.

 

[DominiConnor]

Daniel is right and FPGA is even more hardware dependent.

That's a big reason why I've pushed the idea that to be a good developer you need to understand more of the hardware and what the CPU does with your code.
Of course housekeeping code, like GUIs can be written in Java or VB and data pumping in Java, C++ or a scripting language.

Another point is that you will find it increasingly hard to get by with just one language.

 

[Andy Nguyen]

Another point is that you will find it increasingly hard to get by with just one language.

This is a good one. I notice some MFE students are picking up Python/R/SQL/etc besides their main language of C++.