byteLAKE's expertise on NVIDIA technologies

We have been working on NVIDIA’s platforms starting from Tesla and Fermi architectures. Today, we offer our exceptional experience with the following architectures:

  • Tesla,
  • Fermi,
  • Kepler (i.e. K80 for servers, GeForce GTX Titan for desktop, Jetson for mobile),
  • Maxwell (i.e. NVIDIA GeForce GTX 980 for desktop),
  • Pascal (i.e. P100 for servers),
  • Tesla V100, T4

We are experts in all relevant software stacks (i.e. CUDA, OpenCL, OpenACC) and have successfully delivered numerous projects on NVIDIA technologies.
These are mainly in the areas of:

  • software porting to CPU+GPU architectures,
  • HPC weather simulation optimization (i.e. on Piz Daint),
  • on-device AI inferencing and neural network training.

Deep Learning optimization

Meanwhile, we have also published several research papers using NVIDIA hardware for prestigious magazines i.e. Concurrency and Computation: Practice and Experience, Parallel Computing, Journal of Supercomputing. Find out more details in our presentation on SlideShare and by going through our case studies.

byteLAKE's example public research

NVIDIA logo