NVIDIA freshly released their new flagship Tesla GPU, the Tesla K40. This GPU features more memory, higher clock rates, and more CUDA cores than the previous top-end card, the K20X. But what performance improvements can we expect for financial applications? We’ve put the new card to the test and compared it to the K20X using a Monte-Carlo LIBOR swaption portfolio pricer, a real-world financial algorithm that we’ve already used in other benchmarks.
Xcelerit were represented at last week’s Fixed Income Conference in Munich’s Sofitel Bayerpost. We had a busy display stand in the hotel concourse and our CTO Jorg Lotze gave a presentation revealing some of our CVA implementation secrets. This session was very well attended despite the late dinner at the Augustiner Keller on the evening before.
Intel just released its new Ivy-Bridge server processor line (Xeon E5 v2 series), promising significant performance gains over previous-generation Sandy-Bridge processors. In this post, we will compare the two generations for a financial application – a Monte-Carlo LIBOR Swaption Porfolio pricer.
On September 25th, Xcelerit in association with the Wilmott Forum delivered a briefing on Monte-Carlo Simulations using GPUs. The event was heavily oversubscribed and the audience of quants and other financial industry specialists listened as Dongsheng Lu, Managing Director and Head of Quantitative Research at BNY Mellon explained the complexity and challenges behind CVA and FVA calculations. His presentation was followed by John Ashley of NVIDIA, who explained how GPUs could meet the computational challenges. The event concluded with a presentation by Xcelerit’s Hicham Lahlou, who explained how heavy CVA calculations could be approached in software and how existing code can easily be retargeted to many different accelerator platforms.
On September 26th, Xcelerit’s New York tour continued with a seminar entitled “The Joy of GPUs: Awesome Compute Power for Banks on Budget”. The event was jointly held with IBM and NVIDIA and the venue was IBM’s Wall Street Center of Excellence on Madison Avenue. The seminar was opened by Dave Weber, IBM’s program director at the Wall Street Center. He painted a picture of IBM’s experience deploying advanced server hardware in many different financial industry sites. Jeff Sporn from NVIDIA elaborated on his company’s experience in deploying GPU hardware in a financial setting. Finally Hicham Lahlou from Xcelerit described how to “make Financial Applications Sing on GPUs”.
Following on from our recent blog post detailing our comparison of Intel Xeon Phi with NVIDIA Tesla in financial applications, Xcelerit’s CTO Jorg Lotze was interviewed by HPC wire. You can listen and read about it here
Clearly some of our customers do and we are flattered to be nominated in no less than four categories in the HPCwire Readers Choice Awards including:
- Best use of HPC in financial services
- Best use of HPC in “edge HPC” application
- Top 5 new products or technologies to watch
- Top 5 vendors to watch
HPCwire is the leading publication for news and information from the high performance computing industry and it has become the portal of choice for business and technology professionals from a range of application areas.
Now that we are nominated, we’d like to win – so we are calling on customers, partners, suppliers and anyone who knows what we are all about to Click on Xcelerit
The winners of the awards will be announced at The Supercomputing 2013 (SC’13) conference in Denver, Colorado in November, so get clicking!
Accelerators battle for compute-intensive analytics in Finance
At Xcelerit, people often ask us: “Which is better, Intel Xeon Phi or NVIDIA Kepler?” The general answer has to be “it depends,” as this is heavily application-dependent. But what if we zoom-in on real-world problems in computational finance? The kinds of problems that quants in investment banks and the financial industry are dealing with every day. Let’s analyse two different financial applications and see how they perform on each platform. To cover different types of algorithms often found in finance, we chose an embarrassingly parallel Monte-Carlo algorithm (with full independent paths) for the first test application, and a Monte-Carlo algorithm with cross-path dependencies with iterative time-stepping for the second.
[Update 1-Oct-2013: (American Monte-Carlo application only)]
- Algorithm update and avoiding temporary storage: affects GPU and Xeon Phi heavily, updated the performance numbers
- Updated performance figures for Ivy-Bridge CPU (Xeon E5-2697 v2) and Xeon Phi Processor (Xeon Phi 7120P)
- Replaced absolute times with speed-ups vs. sequential for better readability
Last month, Xcelerit and their partners Excelian and Nvidia held a breakfast briefing to an invited audience in the city of London. The event was hugely popular targeting individuals with IT expertise and giving a snapshot on how far GPU usage has spread within the day-to-day financial services production environment. An overview of the three presentations follows:
More positive coverage of Xcelerit’s achievements appeared this month in Forbes magazine where Tom Groenfeldt’s regular column discussed HSBC’s experience using the Xcelerit SDK for Credit Value Adjustment (CVA) calculations.