GPUs and flash memory are used in numerous finance applications today and the number is rapidly growing. The need to store and transcribe huge amounts of data through data and image processing is becoming overwhelming. The more GPUs and flash memory available, the quicker the data can be used. GPU appliances supporting multiple NVIDIA GPUs and Intel Phi coprocessors and Flash memory appliances supporting multiple Fusion-io flash memory cards are quickly becoming the best and most economical way of accomplishing this tremendous feat.
The financial industry is commonly associated with the phrase, “high-risk, high-reward.” A very small error can cost companies and individuals a lot of money. Even without any errors, there is still a high risk of large losses. The financial industry has kept up with advances in technology and gone are the days of doing calculations by hand. Financial traders today use complex algorithms and powerful computers to help them make decisions. Because the financial industry is a fast paced business, it’s important for traders to use powerful systems that can make calculations quickly and accurately. Monte Carlo simulations are widely used in the finance industry to assess financial investments.
A large bank utilizes the OSS 2U Compute accelerator with NVIDIA K20X GPUs connected to an IBM 3650M4 server to run Monte Carlo Simulations. NVIDIA GPUs can support and speed up applications that run Monte Carlo Simulations, such as Murex MACS Analytics Library and Tanay ZX Lib (Fuzzy) Logix. The ability to create and execute Monte Carlo pricing models much faster allows traders and risk managers to assess alternative modeling scenarios and enhance risk analysis. The improved understanding of the derivative contract and its risk sensitivities increases the profit potential of the deal.
High-frequency trading (HFT) requires sophisticated technology to trade rapidly. High frequency traders move in and out of positions in fractions of a second. When something happens that quickly, it's important to have the fastest technology possible. HFT has been in practice since the late 1990s, where transactions would take several seconds as opposed to the microseconds they now take today.
At the forefront of computing speed right now are GPUs as they provide faster data transfer than CPUs. Many servers today cannot support high end, powerful GPUs because they lack the necessary power and cooling. Imagine an HFT firm using a server with one or two low end GPUs. If the server were attached to the OSS High Density Compute Accelerator with 16 high end GPUs, computations could be sped up exponentially. If high frequency traders make more money by making more trades, then the faster the equipment is, the more money a trader can make. At One Stop Systems, we endeavor to stay on the cutting edge of high performance computing technology. So when even faster GPUs enter the marketplace, you can rely on us to upgrade our compute accelerators to accommodate the fastest technology.
More and more financial firms are looking to change the way they do investment risk analysis and computation. They are moving away from expensive internally developed systems and moving toward newer solutions from experts in the high performance computing market. Financial firms need accurate calculations at the highest possible speed if they want to keep up in the fast paced financial marketplace. The right application software, paired with the right technology can help them achieve that. As technology improves, so do the methods in the financial market for calculating partial differential equations (PDE), probability, numerical analysis, etc. For example, with the Black-Scholes model (one specific PDE), an options market participant can theoretically estimate the price of European-style options.
If you're looking for the fastest computation of partial differential equations, using GPUs is beneficial because they provide faster data transfer than CPUs. One NVIDIA GPU card provides the same compute power as up to 400 CPU cores, while also reducing the hardware footprint drastically. Many servers can accommodate a few GPUs, but not usually high end ones. A high end GPU may use upwards of 300W of power, which is often too much for a server to handle internally. Imagine a financial firm using a server with one or two low end GPUs running Murex MACS Analtyics Library. Now imagine that same server attached to the OSS High Density Compute Accelerator with 16 high end GPUs; Murex would be able to perform partial differential equations exponentially faster.
There are many different types of risks in the financial industry. There's credit risk and market risk, but also, volatility, inflation risks, foreign exchange etc. Because of the advancement of technology, specifically GPUs, you can now calculate these risks very quickly. GPUs allow you to calculate risk in seconds rather than minutes or hours. This allows you to run even more simulations than before, thereby increasing the quality of your calculations. With this increased confidence in your data, you’re more competitive in the financial market. Not only do GPUs speed things up, but they also allow you to run more complex models than ever before. Now you don’t have to wait overnight for your risk calculations and results so you’re able to quickly make better decisions and reduce risk.
There are many applications that allow you to analyze risk. For example, an application like SunGard Adaptiv Analytics would allow you to analyze risk very quickly with NVIDIA GPUs. An NVIDIA Tesla K40 GPU can perform over 60 million valuations per second with SunGard. Now imagine a financial company running a server that’s connected to the OSS High Density Compute Accelerator with 16 NVIDIA Tesla K40 GPUs. And GPU technology is advancing at an incredible pace. At One Stop Systems, we keep up the pace in order to bring you the fastest technology possible.
Here are some computational finance applications that can be sped up by using GPUs:
Our compute accelerators support from one to sixteen double-wide PCIe cards and can be cabled up to four host computers through PCIe x16 Gen3 connections each operating at 128Gb/s. The all-steel construction chassis house power supplies, fans, and a system monitor that monitors the fans, temperature sensors and power voltages. Front panel LEDs signal minor, major or critical alarms. The compute accelerators are transparent and do not require software except for the drivers required by the PCIe add-in cards. Compute accelerators are the best appliance for applications that require a large amount of compute power.Learn More
Our flash storage arrays support from one to thirty-two single-wide PCIe cards and can be cabled up to four host computers through PCIe x16 Gen3 connections each operating at 128Gb/s. The light-weight, compact chassis house power supplies, fans, and a system monitor that monitors the fans, temperature sensors and power voltages. Flash storage arrays are the best appliance for applications that require a large amount of fast, flexible storage.
To read an independent review of the 3U FSA done by Tom's IT Pro, click here.Learn More