Forums  > Software  > CUDA knowledge - does it really increase employee's value?  
     
Page 1 of 1
Display using:  

finanzmaster


Total Posts: 168
Joined: Feb 2011
 
Posted: 2011-02-01 10:34
Hello together,

There was pretty much talks (hype?) on CUDA, there are some topics at this forum as well.
But how widely does one engage CUDA in financial world?
Classical example is BNP Paribas.
But as to quant jobs in Germany I, for one, have seen [so far] only one job, where CUDA knowledge was desired (it was not even a must-have).

Probably, there is more demand for CUDA in City or NY?

www.yetanotherquant.com - Knowledge rather than Hope: A Book for Retail Investors and Mathematical Finance Students

FDAXHunter
Founding Member

Total Posts: 8372
Joined: Mar 2004
 
Posted: 2011-02-01 10:43
Dude.. CUDA is just one small tool in a very large toolbox. Not knowing CUDA will not hurt you. Just as knowing one more language will not really help you.

The Figs Protocol.

finanzmaster


Total Posts: 168
Joined: Feb 2011
 
Posted: 2011-02-01 11:39
I would not oversimplify so much.
CUDA is not yet another programming language, it is a paradigm (of programming parallel GPUs).
I mean the following: when one comes from C to C++, one learns not a new language but the new concepts: OOP, template programming - and it is much more.

Moreover, CUDA can increase computational performance upto 100 times, so it is not "just one small tool", it can be a breakthrough.
But does anyone really need it? In high-frequency trading very likely.
On the other hand, if a bank must satisfy a regulator and the latter says 1000 Scenario-simulations will do, why should a bank bother trying to perform 100000 simulations in the same time?

So I am curious how many employers do really care. As I said, in Germany there are not many. So what about the City or NY?

www.yetanotherquant.com - Knowledge rather than Hope: A Book for Retail Investors and Mathematical Finance Students

FDAXHunter
Founding Member

Total Posts: 8372
Joined: Mar 2004
 
Posted: 2011-02-01 12:06
I'm telling you, it is one more tool.

The Figs Protocol.

pj


Total Posts: 3457
Joined: Jun 2004
 
Posted: 2011-02-01 12:12
it is not a tool it's a paradigm Cool

вакансия "Программист Психологической службы" -але! у нас ошибко! не работает бля-бля-бля -вы хотите об этом поговорить?

FDAXHunter
Founding Member

Total Posts: 8372
Joined: Mar 2004
 
Posted: 2011-02-01 12:20
Hammertime <--- Tool, you see?

The Figs Protocol.

pj


Total Posts: 3457
Joined: Jun 2004
 
Posted: 2011-02-01 12:51
Piss <---paradigm.
did you get it?

вакансия "Программист Психологической службы" -але! у нас ошибко! не работает бля-бля-бля -вы хотите об этом поговорить?

prophet
Banned

Total Posts: 149
Joined: Oct 2004
 
Posted: 2011-02-01 18:45
CUDA can increase computational performance upto 100 times

For highly vectorizable algorithms with fairly regular memory access patterns, yes. For most algorithms it depends on many factors. Many algorithms can not be easily mapped to a GPU. One must also consider the implementation cost of CUDA versus out-of-the-box GPU accelerator tools like www.accelereyes.com

CUDA is just basic extensions upon C. A good coder can master it in a few days. The real skill of interest is understanding hardware (GPU vs CPU), understanding the problem and choosing the right tools (CUDA, OpenCL, C/C++, etc) to map the problem to computational hardware at appropriate cost.


gary_badridge


Total Posts: 6
Joined: Jan 2011
 
Posted: 2011-02-04 02:37
It's better to have a very solid grasp of the high performance languages (c/c++) and be familiar with how the problems can be parallelized or broken down (ie fluids vs for example search). High-end hardware changes very quickly, but the basics stay the same.

spaghetti_o_o


Total Posts: 21
Joined: Sep 2010
 
Posted: 2011-03-04 03:15
You can run a search on dice.com or bloomberg for "cuda finance" and see what gives, then run the same with "c++ linux finance". Or ask your favorite recruiter.

In the end it's always harder to design, implement and *debug* a robust parallel solution then to learn a specific technology or api. Also keep in mind that today you can get a 12 cores server for ~$3,000 (24 cores if hyper-threaded) and 1Gb interconnects are cheap, so it's not like we are missing parallel platform solutions.

chiral3
Founding Member

Total Posts: 5100
Joined: Mar 2004
 
Posted: 2011-03-04 13:55
also look into thrust to counter some of the c-ness.

Nonius is Satoshi Nakamoto. 物の哀れ

finanzmaster


Total Posts: 168
Joined: Feb 2011
 
Posted: 2019-05-11 20:11
It is interesting to recall this post after more than 8 years.
So, guys, I was right: CUDA turned out to be not just yet another programming language, it turn out to bea paradigm.

However, not in quantitative finance (which are not anymore so sexy as before) but in AI/ML hype.

BTW, in 2011 I bought NVIDIA stock but sold them prematurely with a modest profit.
Later in 2016 I expected that the NVIDIA stock will halve... and it recently did... but by that time I was the short was also premature :)
https://letyourmoneygrow.com/2016/12/25/my-put-option-on-nvidia-a-case-study-of-nearly-perfect-trading-decision/

www.yetanotherquant.com - Knowledge rather than Hope: A Book for Retail Investors and Mathematical Finance Students

rickyvic


Total Posts: 188
Joined: Jul 2013
 
Posted: 2019-05-14 13:09
It is just not that much advantageous with our day to day stuff.
I never find the speed issue that big once I optimise the code and use a speedy machine that I have to look for a way to push it to a gpu.

For real time I heared some people use it instead of fpgas.

I would love to use it for large time series manipulations which are the time consuming parts but then again if I spend time on that recoding I'd better spend it rewriting optimised cpu code.

"amicus Plato sed magis amica Veritas"
Previous Thread :: Next Thread 
Page 1 of 1