high performance computing on graphics processing units: hgpu.org

Deep learning framework supported by CUDA parallel computing platform boosts advances of studies on machine learning. The advantage of parallel processing largely comes from an efficiency of matrix-matrix multiplication using many CUDA-enabled graphics processing units (GPU). Therefore, for recurrent neural networks (RNNs), the usage of a zero-filled matrix representing variable lengths of sentences for a […]

We report on an extension to the C-Reduce tool, for automatic reduction of C test cases, to handle OpenCL kernels. This enables an automated method for detecting bugs in OpenCL compilers, by generating large random kernels using the CLsmith generator, identifying kernels that yield result differences across OpenCL platforms and optimisation levels, and using our […]

Programming today’s increasingly complex heterogeneous hardware is difficult, as it commonly requires the use of data-parallel languages, pragma annotations, specialized libraries, or DSL compilers. Adding explicit accelerator support into a larger code base is not only costly, but also introduces additional complexity that hinders long-term maintenance. We propose a new heterogeneous compiler that brings us […]

We present the systematic design of a testing environment that uses stressing and fuzzing to reveal errors in GPU applications that arise due to weak memory effects. We evaluate our approach on seven GPUs spanning three Nvidia architectures, across ten CUDA applications that use fine-grained concurrency. Our results show that applications that rarely or never […]

To concisely and effectively demonstrate the capabilities of our program transformation system Loo.py, we examine a transformation path from two real-world Fortran subroutines as found in a weather model to a single high-performance computational kernel suitable for execution on modern GPU hardware. Along the transformation path, we encounter kernel fusion, vectorization, prefetching, parallelization, and algorithmic […]

In the context of this thesis, our focus is on numerical linear algebra, more precisely on solution of large sparse systems of linear equations. We focus on designing efficient parallel implementations of MaPHyS, an hybrid linear solver based on domain decomposition techniques. First we investigate the MPI+threads approach. In MaPHyS, the first level of parallelism […]

There are billions of lines of sequential code inside nowadays’ software which do not benefit from the parallelism available in modern multicore architectures. Automatically parallelizing sequential code, to promote an efficient use of the available parallelism, has been a research goal for some time now. This work proposes a new approach for achieving such goal. […]

Optimal performance is an important goal in compute intensive applications. For GPU applications, this requires a lot of experience and knowledge about the algorithms and the underlying hardware, making them an ideal target for autotuning approaches. We present an auto-tuner which optimizes array layouts in CUDA applications. Depending on the data and program parameters, kernels […]

Frequent subgraph mining (FSM) is an important problem in numerous application areas, such as computational chemistry, bioinformatics, social networks, computer programming languages, etc. However, the problem is computationally hard because it requires enumerating possibly an exponential number of candidate subgraph patterns, and checking their presence in a single large graph or a database of graphs. […]

With increasing core-count, the cache demand of modern processors has also increased. However, due to strict area/power budgets and presence of poor data-locality workloads, blindly scaling cache capacity is both infeasible and ineffective. Cache bypassing is a promising technique to increase effective cache capacity without incurring power/area costs of a larger sized cache. However, injudicious […]

Due to better parallel density and power efficiency, GPUs have become more popular for use in scientific applications. Many of these applications are based on the ubiquitous Message Passing Interface (MPI) programming paradigm, and take advantage of non-contiguous memory layouts to exchange data between processes. However, support for efficient non-contiguous data movements for GPU-resident data […]

The alpaka library defines and implements an abstract hierarchical redundant parallelism model. This model exploits parallelism and memory hierarchies on a node at all levels available in current hardware. This allows to achieve portability of performant codes across various types of accelerators by ignoring specific unsupported levels and utilizing only the ones supported on a […]