Cognitive radio researchers at Trinity College, Dublin have turned to Xcelerit software for help in a plan to build future phone networks that can organize themselves. Traditionally, cellular phone companies like Vodafone, O2 and T-mobile bid against each other for exclusive access to radio 'spectrum'. It is a slow and expensive process that often ends up with spectrum being under used. The research team at the CTVR telecommunications research centre based at Trinity College Dublin are exploring ways to avoid the need to carve up spectrum in advance.
The researchers are employing 'Game Theory' -- "a kind of analysis associated with the brilliant mathematician John Nash, whose story was the topic of the Hollywood movie 'A Beautiful Mind'," states Prof. Luiz DaSilva, leader of the CTVR research team. Anyone wanting to use a radio or mobile phone in a space can compete for the right to use the spectrum eventually reaching a "Nash Equilibrium" where each of their needs is adequately met. The research team came up with the right machine learning algorithms, but to put them to the test, they needed to build a simulation model employing game theory to show that it would work in practice. "To test our algorithms, we wanted to simulate a reasonably sized network and try out many different combinations of transmitters and receivers to be sure that equilibrium is reached in all cases," said researcher Dr. Irene Macaluso. "A 'reasonably sized network' means individual testing of many hundreds of thousands of cases and our typical analyses could take several days using sequential programs."
The CTVR team turned to Xcelerit, a company dedicated to making the so-called many-core computer technology accessible to mainstream programmers. Over the last few years, processor makers have shifted from increasing the speed of individual processor cores to adding more and more cores on a single chip (CPU). Further, the immense computational power of graphics cards (GPUs) has been made available for general purpose programming. However, programmers need to change their software considerably to take advantage of these processors and that's where Xcelerit comes in. "Programmers cry out for ways to simplify programming these processors" comments Xcelerit's CEO Hicham Lahlou. "Our software development kit (SDK) splits the tasks up automatically to keep all processor cores busy. And the beauty is that the SDK ensures that a single codebase can run on multi-core CPUs, GPUs, and any combination of these in a cluster."
"We were delighted to put our software toolkit to work on problems in game theory" comments Xcelerit's Lahlou. "The CTVR team was able to make minor changes to their simulator implementation to make it work with our Xcelerit SDK. Once that was done, they could run it on different machine configurations including GPU hardware from Nvidia®." CTVR did not have this specific hardware to hand, but were able to source it easily from PEER 1 Hosting who has recently launched a GPU Cloud Service where access to high performance servers and GPUs can be rented on demand.
The results were quite dramatic: On a PEER 1 system with two Intel® Xeon® E5620 CPUs it was possible to speed up the code by a factor of 13 compared to a sequential implementation using a single CPU core. When two Nvidia® Tesla™ M2050 GPUs were added, speedups of 140x were recorded without further code changes. "This meant that some of our simpler computations were completing in under a second and our typical simulations went from 35 hours down to just 15 minutes," said Dr. Macaluso. "Sourcing the extra compute horse-power from the cloud was great because we had no lead time for new hardware and we can rent it whenever we need it in future," she said.
Xcelerit is planning to deploy its toolkit widely in education and research. "Universities and research centres are great customers for us," says Hicham Lahlou, "They really stretch our product's capabilities and are very influential when it comes to industry take-up."
Cite This Page: