high performance computing on graphics processing units: hgpu.org

High-performance tensor programs are crucial to guarantee efficient execution of deep learning models. However, obtaining performant tensor programs for different operators on various hardware platforms is notoriously difficult. Currently, deep learning systems rely on vendor-provided kernel libraries or various search strategies to get performant tensor programs. These approaches either require significant engineering efforts in developing […]

We apply the SYCL programming model to the Rodinia benchmark suite, describe the transformations from the OpenCL implementations to the SYCL implementations, and evaluate the benchmarks on microprocessors with a CPU and an integrated GPU. The publicly available implementations of the benchmark suite will track the development of the SYCL compilers, and provide programs for […]

Compilers play a pivotal role in the software development process for microprocessors, by automatically translating high-level programming languages into machinespecific executable code. For a long time, while processors were scalar, compilers were regarded as a black box among the software community, due to their successful internal encapsulation of machine-specific details. Over a decade ago, major […]

The requirements and responsibilities assumed by software have increasingly rendered it to be large and complex. Testing to ensure that software meets all its requirements and is free from failures is a difficult and time-consuming task that necessitates the use of large test suites, containing many test cases. Time needed to compile and execute large […]

We propose a universal method for the evaluation of generalized standard materials that greatly simplifies the material law implementation process. By means of automatic differentiation and a numerical integration scheme, AutoMat reduces the implementation effort to two potential functions. By moving AutoMat to the GPU, we close the performance gap to conventional evaluation routines and […]

The time and effort involved in hand-designing deep neural networks is immense. This has prompted the development of Neural Architecture Search (NAS) techniques to automate this design. However, NAS algorithms tend to be extremely slow and expensive; they need to train vast numbers of candidate networks to inform the search process. This could be remedied […]

The execution of agent-based simulations (ABSs) on hardware accelerator devices such as graphics processing units (GPUs) has been shown to offer great performance potentials. However, in heterogeneous hardware environments, it can become increasingly difficult to find viable partitions of the simulation and provide implementations for different hardware devices. To automate this process, we present OpenABLext, […]

Recently, OpenCL has been emerging as a programming model for energy-efficient FPGA accelerators. However, the state-of-the-art OpenCL frameworks for FPGAs suffer from poor performance and usability. This paper proposes a highlevel synthesis framework of OpenCL for FPGAs, called SOFF. It automatically synthesizes a datapath to execute many OpenCL kernel threads in a pipelined manner. It […]

The fundamental operation of matrix multiplication is ubiquitous across a myriad of disciplines. Yet, the identification of new optimizations for matrix multiplication remains relevant for emerging hardware architectures and heterogeneous systems. Frameworks such as OpenCL enable computation orchestration on existing systems, and its availability using the Intel High Level Synthesis compiler allows users to architect […]

Multiplication of a sparse matrix to a dense matrix (SpDM) is widely used in many areas like scientific computing and machine learning. However, existing works under-look the performance optimization of SpDM on modern many-core architectures like GPUs. The storage data structures help sparse matrices store in a memory-saving format, but they bring difficulties in optimizing […]

This paper investigates the multi-GPU performance of a 3D buoyancy driven cavity solver using MPI and OpenACC directives on different platforms. The paper shows that decomposing the total problem in different dimensions affects the strong scaling performance significantly for the GPU. Without proper performance optimizations, it is shown that 1D domain decomposition scales poorly on […]

SYCL is a parallel programming model for developing single-source programs for running on heterogeneous platforms. To this end, it allows for one code to be written which can run on a different architectures. For this study, we develop applications in SYCL which are representative of those often used in High-Performance Computing. Their performance is benchmarked […]