by Alain Tiquet

Periodically, we’re using this blog to profile some of the companies that participated in NVIDIA’s Emerging Companies Summit. You can learn more about innovative companies that use NVIDIA’s GPU technology in the GPU Ventures Zone.

When it comes to making money in the financial markets, the ability to see opportunities is key. While there are many existing tools that extract data for users, the tools don’t always present data in a meaningful way. Using CUDA parallel computing and GPUs, Houston-based startup Aqumin has developed a system that takes complex financial information and presents it intuitively, visually and in 3D. They find needles in a haystack and hand them over to users.

For financial information to be useful it needs to be shown in relation to other data. You don’t know if a company’s P/E ratio is good or bad unless you can compare it across a list of similar companies – its neighbors. Aqumin’s AlphaVision technology literally presents a neighborhood, showing financial data as information in a cityscape.

By presenting individual securities as color-coded 3D buildings on city blocks, AlphaVision enables analysts and traders to interpret and interact with huge amounts of data in an environment that feels familiar and manageable. The size and color of a building and its neighbors can tell an analyst key information – such as market capitalization, price difference, etc. – at a glance. From there, an AlphaVision customer can easily drill down into more detail or interact with the cityscape to perform comparative analyses, complex financial operations and searches.

“Financial information has been 2D,” Aqumin CTO Sean Spicer told NVIDIA VP of Business Development Jeff Herbst when the two met for a video chat at ECS. With Aqumin, “when people think about how to consume financial information, they’ll be starting in a 3D landscape.”

For financial analysts, who spend the bulk of their time either gathering or analyzing financial data, AlphaVision can significantly speed the development of their financial models. And while an individual trader without AlphaVision can manage roughly 50-60 securities at a given time, with AlphaVision, he or she can manage many more times that amount. (In his ECS presentation, Spicer showed a monitor screenshot from one multitasking trader who was managing 800 positions.)

Aqumin uses GPUs in two different ways. Behind the scenes, its algorithms have been accelerated with parallel computing using the CUDA architecture and NVIDIA Tesla GPUs. On the customer side, Aqumin relies on NVIDIA Quadro and GeForce GPUs for graphics quality.

Although globally the financial sector has fallen on hard times, Aqumin is seeing opportunity in AlphaVision’s elegant datascapes. They announced a deal with Bloomberg in September 2009 that enables Terminal users worldwide to run its software and recently announced a license agreement with the Chicago Board Options Exchange.