Supercomputing at your fingertips

There are a wide range of scientific, financial and research applications beginning to take advantage of the computing performance of the new generation of Graphics Processing Units (GPUs) on modern graphics cards. Modern GPUs have up to 2,048 processing cores that can run jobs in parallel.

GenoKey’s tools are based on a novel approach combining geometry with formal logic in a mathematical technique called Array Based Logic. This technique reduces complex problem spaces to dense nested arrays which are ideal for massive parallelization on low cost computing devices such as GPUs. GenoKey’s algorithms work up to 200x faster on a GPU than an equivalently priced CPU due to their efficient exploitation of array operations on the massively parallel architecture. This allows GenoKey’s tools to operate with dataset sizes and complexities that massively exceed existing applications. Running GenoKey’s analytical algorithms on 2,048 GPU cores simultaneously, puts the power of a supercomputer at your disposal for a tiny fraction of the capital and operational cost.

Complex studies such as Genome Wide Association Studies can be run in minutes on your desktop PC rather than months on a supercomputing cluster as shown below. The benchmark data below shows the time taken for a GWAS analysis of 3-SNP combinations across 2,000 patients and 2,000 controls each with a 479,000 SNP genotype. The times include the time required to validate the SNP combinations by computing significant 3-SNP combinations for 1,000 random permutations of ‘patients’ and ‘controls’.

New GWAS Benchmark

1 NVIDIA GTX590 Card with 960 1.2 GHz GPU cores
2 Laptop computer with dual core 3 GHz CPU
3 Based on data provided by major research centre running broadly similar diabetes-I study
4 Extrapolation based on number of combinations required

Comments are closed.