When Bloomberg was looking to upgrade its server farm the company found out that NVIDIA Tesla based servers that take advantage of GPGPU computing are much more cost-effective for this job than traditional servers. The 48 server/GPU pairs are just as powerful as 1,000 servers with eight cores each, while using just one-third of the energy and occupying significantly less data center space.
Bloomberg and nVidia engineers worked together to get the pricing software to run on the GPUs. "The underlying math and algorithms are proprietary to Bloomberg," says Andy Keane, general manager, Tesla supercomputing at nVidia. "We provide training, expertise to make the Bloomberg software GPU-compatible. There's a bit of a wall between the two to protect Bloomberg's intellectual property." Rewriting, restructuring and testing the code to run over the GPUs took about a year. "This service is mission-critical to our customers, they rely on it to make decisions, so we had an extensive testing period," Edwards says.
Part of the pricing application, data gathering, doesn't lend itself well to GPU computing, Edwards notes, because it can't be parallelized. The x86 servers also prepare the problems to be parallelized. But about 90% of the work does run on the GPU platform, he says.