high performance computing on graphics processing units: hgpu.org

The numerical solution of the Vlasov-Fokker-Planck equation is a well established method to simulate the dynamics, including the self-interaction with its own wake field, of an electron bunch in a storage ring. In this paper we present Inovesa, a modularly extensible program that uses OpenCL to massively parallelize the computation. It allows a standard desktop […]

An increasingly large number of HPC systems rely on heterogeneous architectures combining traditional multi-core CPUs with power efficient accelerators. Designing efficient applications for these systems has been troublesome in the past as accelerators could usually be programmed using specific programming languages threatening maintainability, portability and correctness. Several new programming environments try to tackle this problem. […]

Connected component labeling (CCL) is a traditionally sequential problem that is hard to parallelize. This report aims to test the performance of solving CCL using massively parallel hardware through GPGPU. To achieve this several CCL algorithms were researched and implemented using C++ and OpenCL. The results showed an improvement of up to a factor of […]

Buffer overflows are a common source of program crashes, data corruption, and security problems. In this work, we demonstrate that GPU-based workloads can also cause buffer overflows, a problem that was traditionally ignored because CPUs and GPUs had separate memory spaces. Modern GPUs share virtual, and sometimes physical, memory with CPUs, meaning that GPU-based buffer […]

A single CPU reached its limit of computational throughput over a decade ago, and as a response the technology industry was forced to shift to parallel processing. Today processors are increasingly parallel, with increasing core counts, wider SIMD lanes, and more hardware threads. Systems are also heterogeneous, so that a single workstation, server, or smartphone […]

As Central Processing Units (CPUs) and Graphical Processing Units (GPUs) get progressively better, different approaches and designs for implementing algorithms with high data load must be studied and compared. This work compares several different algorithm designs and parallelization APIs (such as OpenMP, OpenCL and CUDA) for both CPU and GPU platforms. We used the Cholesky […]

Predictive modeling using machine learning is an effective method for building compiler heuristics, but there is a shortage of benchmarks. Typical machine learning experiments outside of the compilation field train over thousands or millions of examples. In machine learning for compilers, however, there are typically only a few dozen common benchmarks available. This limits the […]

Programming languages such as C for CUDA, OpenCL or ISPC have contributed to increase the programmability of SIMD accelerators and graphics processing units. However, these languages still lack the flexibility offered by lowlevel SIMD programming on explicit vectors. To close this expressiveness gap while preserving performance, this paper introduces the notion of Call Re-Vectorization (CREV). […]

This report describes the physical concept of fluids as well as a mathematical model for fluids governed by the Navier-Stokes equations. The Smoothed Particle Hydrodynamics method (SPH) for simulating fluids is described, and implementation details of the method are explained. Numerical integration methods such as Euler and Leap-Frog integration are discussed. The presented result is […]

This paper reports on the results of a study to accelerate string tokenization using FPGAs suitable for both IoT gateways and data center servers. The prototype developed with Xilinx High-Level Synthesis software runs at 200 MHz and processes up to 32 ASCII characters per clock cycle. It incorporates either OpenCL or our own framework (Volvox) […]

Image and video compression play a major role in the world today, allowing the storage and transmission of large multimedia content volumes. However, the processing of this information requires high computational resources, hence the improvement of the computational performance of these compression algorithms is very important. The Multidimensional Multiscale Parser (MMP) is a pattern-matching-based compression […]

In recent years, deep neural networks (DNNs), have yielded strong results on a wide range of applications. Graphics Processing Units (GPUs) have been one key enabling factor leading to the current popularity of DNNs. However, despite increasing hardware flexibility and software programming toolchain maturity, high efficiency GPU programming remains difficult: it suffers from high complexity, […]