Hardware

Oneye’s ATS solution is cleverly designed to take advantage of the costly systems that a finance production environment already has in place. The key to our business success is in the cost effectiveness of GPU hardware Integration.

Using Nvidia’s GPU Telsa technology, our solutions integrate GPU technology through application of NVIDIA's CUDA programming language. GPU hardware offers Complete flexibility and scalability, its Multi thread parallel processing technology means that financial markets are recording processing speedup times of 100!

Scalability of ATS Hardware Solutions

Instant Aceleration

GPU Acceleration Board

NVIDIA’s GPU Tesla processing board is a dedicated computing board that is suited for a single PC or workstation. The GPU Computing Processor transforms a workstation into a high-performance computing machine.

Tesla processing boards deliver today up to a TeraFlop of parallel computation.

The GPU computing processor can be used in existing systems, partnered with high-performance CPUs.

Small Scale Operations

Lightweight GPU Station

The lightweight GPU solution offers adaptation of the common desktop PC to enable High Performance Computing capabilities.

This system offers a small margin of scalability by enabling the user to add GPU boards to it. This is ideal for the individual trader.

Medium scale operations

GPU dedicated server

Small to Medium size financial firms can benefit from a small scale desk-side GPU integrated server. The leap from the GPU adapted lightweight station to an actual GPU integrated server offers a process that is much faster and more energy-efficient than a shared cluster in a data center.

Data Center Deployment

Large Scale GPU Computing System

Large firms or hedge funds can benefit from GPU equiped datacenters. Clouds of NVIDIA 1U servers are ideal for this function. Each server houses up to eight GPU boards, each capable of 1 Teraflops. Meaning that in combination with a market feed appliance any changes in market values, a GPU enhanced server can recompute in the microsecond range. Nvidia is currently developing a second generation of GPU boards that will reportedly cut recomputing time in half again.

Client Benefits

  • With 240 cores per processor, GPU Telsa boards offer a quantum leap in capabilities. With Concurrent execution of thousands of computing threads, financial institutions can get accurate answers, faster. See further reading speedup test on Option pricing models.

  • “High density computing” allows technical professionals and data center managers to deliver more performance without new demands on the electrical and thermal capabilities.

  • This gives green-conscious organizations an excellent choice to achieve their desired level of computation without increasing energy costs.