Once again, I am finding various technical challenges with my proposed HFT platform. I think I have found the ultimate end target which is near zero latency. I will only be detailing my process to my QuantLabs.net Premium Members how to do it. Now is the time as this highly prized technique becomes more valuable to join while it is very affordable.
I have documented my challenges of Matlab Coder and GPU/CUDA technologies. There is a highly powerful solution to all that so read on.
I have been looking at how to implement non supported Matlab Coder toolbox functions as a replacement for a C++ based HFT platform. Here is what I found:
1. Matlab 2012b Coder Toolbox supported function list is not that impressive for the big bucks they charge
2. My C++ Demo video of using gretl, gsl, and other open source libraries Matlab Code does not support
As with any evolution of learning new technologies, there are unknown limitation with those that you choose. As a result, here are the challenges of using technologies like GPU, OpenGL, and CUDA:
1. Can GPU CUDA be used for real time data analysis for HFT?
2. Bye bye GPGPU and GPU? This Nvidia CUDA vs FPGA debate for real time data HFT systems
So is FPGA the true answer to all these problems? This link below is the most important of all!
1. For HFT: FPGA notes with potential capabilities thanks to Matlab’s Simulink and HDL Code Generation
In tomorrow’s email. I will document the benefits of using this unknown technology and amazing benefits you get with it. In the meantime, start preparing your QuantLabs.net Premium Membership because I will be on the only person documenting it online for the world of trading.
There appears to be a huge demand to learn FPGA in trading as we reached our highest second day in traffic after announcing this discovery. As a they say about typical supply and demand, this training is extremely expensive. My learning appears to be the most affordable way to get to market using FPGA.
I will definitely be talking about the evolution of my learning from clusters to FPGA in my private Premium Membership come Dec 18.
It is still affordable but will be going up substantially first week of January, 2013.
Thanks BryanNOTE I now post my TRADING ALERTS into my personal FACEBOOK ACCOUNT and TWITTER. Don't worry as I don't post stupid cat videos or what I eat!