Emerging multi-core architectures such as Intel Xeon are seeing widespread adoption in current and next-generation HPC systems due to their power/performance ratio. Similarly, the recent surge of Deep Learning (DL) models and applications can be attributed to the rise in computational resources, availability of large-scale datasets, and easy to use DL frameworks like Tensorflow, Caffe and...
For writing a new scientific application, portability across existing and future hardware should be the major design goal, as there is a multitude of different compute devices, and codes typically outlive systems by far. Unlike other programming models that address parallelism or heterogeneity, OpenCL does provide practical portability across a wide range of HPC-relevant architectures, and has...
Last year we gave a survey like presentation on our search for a way to realize OpenMP to FPGA offloading and asked for ideas of the community. This is where we got the decisive input we needed to reach our goal.
In this lighting talk we will present the results of it, a first OpenMP to FPGA offloading prototype. It makes use of the LLVM front-end clang for the outlining task and the Intel...
Main goal of the presentation/tutorial is to provide audience an information about key aspects of new generation of Intel MPI product and the way the library may help to HPC/ML workloads.
You will learn about:
- Main difference between old and new generations of the product and rationale behind the changes
- New unique features for multithreading like multiple endpoint support and new...
