Programming Graphics Processing Units Research Articles

Don't forget about synchronization! Guidelines for using locks on graphics processing units

SummaryHeterogeneous devices are becoming necessary components of high performance computing infrastructures, and the graphics processing unit (GPU) plays an important role in this landscape. Given a problem, the established approach for exploiting the GPU is to design solutions that are parallel, without data dependencies. These solutions are then offloaded to the GPU's massively parallel capability. This design principle often leads to developing applications that cannot maximize GPU hardware utilization. The goal of this article is to challenge this common belief by empirically showing that allowing even simple forms of synchronization enables programmers to design solutions that admit conflicts and achieve better performance. Our experience shows that lock‐based solutions to the k‐means clustering problem, implemented using two well‐known locking strategies, outperform the well‐engineered and parallel KMCUDA on both synthetic and real datasets; with an average 8× faster runtimes across all locking algorithms on a synthetic dataset and 1.7× faster on a real world dataset across all locking algorithms (and max speedups of 71.3× and 2.75×, respectively). We validate these results using a more sophisticated clustering algorithm, namely fuzzy c‐means and summarize our findings by identifying three guidelines to help make concurrency effective when programming GPU applications.

Lattice QCD simulations using the OpenACC platform

In this article we will explore the OpenACC platform for programming Graphics Processing Units (GPUs). The OpenACC platform offers a directive based programming model for GPUs which avoids the detailed data flow control and memory management necessary in a CUDA programming environment. In the OpenACC model, programs can be written in high level languages with OpenMP like directives. We present some examples of QCD simulation codes using OpenACC and discuss their performance on the Fermi and Kepler GPUs.

Accelerated Combinatorial Optimization using Graphics Processing Units and C++ AMP

In the course of less than a decade, Graphics Processing Units (GPUs) have evolved from narrowly scoped application specific accelerators to general-purpose parallel machines capable of accommodating an ever-growing set of algorithms. At the same time, programming GPUs appears to have become trapped around an attractor characterised by ad-hoc practices, non-portable implementations and inexact, uninformative performance reporting. The purpose of this paper is two-fold, on one hand pursuing an in-depth look at GPU hardware and its characteristics, and on the other demonstrating that portable, generic, mathematically grounded programming of these machines is possible and desirable. An agent-based meta-heuristic, the Max-Min Ant System (MMAS), provides the context. The major contributions brought about by this article are the following: (1) an optimal, portable, generic-algorithm based MMAS implementation is derived; (2) an in-depth analysis of AMD's Graphics Core Next (GCN) GPU and the C++ AMP programming model is supplied; (3) a more robust approach to performance reporting is presented; (4) novel techniques for raising the abstraction level without sacrificing performance are employed. This represents the first implementation of an algorithm from the Ant Colony Optimisation (ACO) family using C++ AMP, whilst at the same time being one of the first uses of the latter programming environment.

Teaching parallelism with GPUS and a Game of life assignment

In this tutorial, we share our experiences teaching parallelism with GPUs and a Game of Life assignment. The rise of multicore computer hardware (now even in phones and tablets) has introduced new urgency to learn parallel programming. Whereas threads and/or MPI have been taught for many years, programming GPUs (Graphics Processing Units) with hundreds of cores is an exciting new option. With the introduction of languages such as CUDA for general-purpose GPU programming, we can now parallel program GPUs for computational tasks and achieve orders of magnitude improvement in performance.

Sponge

Graphics processing units (GPUs) provide a low cost platform for accelerating high performance computations. The introduction of new programming languages, such as CUDA and OpenCL, makes GPU programming attractive to a wide variety of programmers. However, programming GPUs is still a cumbersome task for two primary reasons: tedious performance optimizations and lack of portability. First, optimizing an algorithm for a specific GPU is a time-consuming task that requires a thorough understanding of both the algorithm and the underlying hardware. Unoptimized CUDA programs typically only achieve a small fraction of the peak GPU performance. Second, GPU code lacks efficient portability as code written for one GPU can be inefficient when executed on another. Moving code from one GPU to another while maintaining the desired performance is a non-trivial task often requiring significant modifications to account for the hardware differences. In this work, we propose Sponge, a compilation framework for GPUs using synchronous data flow streaming languages. Sponge is capable of performing a wide variety of optimizations to generate efficient code for graphics engines. Sponge alleviates the problems associated with current GPU programming methods by providing portability across different generations of GPUs and CPUs, and a better abstraction of the hardware details, such as the memory hierarchy and threading model. Using streaming, we provide a write-once software paradigm and rely on the compiler to automatically create optimized CUDA code for a wide variety of GPU targets. Sponge's compiler optimizations improve the performance of the baseline CUDA implementations by an average of 3.2x.

Graphics processing unit (GPU) programming strategies and trends in GPU computing

Over the last decade, there has been a growing interest in the use of graphics processing units (GPUs) for non-graphics applications. From early academic proof-of-concept papers around the year 2000, the use of GPUs has now matured to a point where there are countless industrial applications. Together with the expanding use of GPUs, we have also seen a tremendous development in the programming languages and tools, and getting started programming GPUs has never been easier. However, whilst getting started with GPU programming can be simple, being able to fully utilize GPU hardware is an art that can take months or years to master. The aim of this article is to simplify this process, by giving an overview of current GPU programming strategies, profile-driven development, and an outlook to future trends.

Real-Time Simulation of Aeolian Sand Movement and Sand Ripple Evolution: A Method Based on the Physics of Blown Sand

Simulation and visualization of aeolian sand movement and sand ripple evolution are a challenging subject. In this paper, we propose a physically based modeling and simulating method that can be used to synthesize sandy terrain in various patterns. Our method is based on the mechanical behavior of individual sand grains, which are widely studied in the physics of blown sand. We accounted significant mechanisms of sand transportation into the sand model, such as saltation, successive saltation and collapsing, while simplified the vegetation model and wind field model to make the simulation feasible and affordable. We implemented the proposed method on the programming graphics processing unit (GPU) to get real-time simulation and rendering. Finally, we proved that our method can reflect many characteristics of sand ripple evolution through several demonstrations. We also gave several synthesized desert scenes made from the simulated height field to display its significance on application.

Providing Source Code Level Portability Between CPU and GPU with MapCG

Graphics processing units (GPU) have taken an important role in the general purpose computing market in recent years. At present, the common approach to programming GPU units is to write GPU specific code with low level GPU APIs such as CUDA. Although this approach can achieve good performance, it creates serious portability issues as programmers are required to write a specific version of the code for each potential target architecture. This results in high development and maintenance costs. We believe it is desirable to have a programming model which provides source code portability between CPUs and GPUs, as well as different GPUs. This would allow programmers to write one version of the code, which can be compiled and executed on either CPUs or GPUs efficiently without modification. In this paper, we propose MapCG, a MapReduce framework to provide source code level portability between CPUs and GPUs. In contrast to other approaches such as OpenCL, our framework, based on MapReduce, provides a high level programming model and makes programming much easier. We describe the design of MapCG, including the MapReduce-style high-level programming framework and the runtime system on the CPU and GPU. A prototype of the MapCG runtime, supporting multi-core CPUs and NVIDIA GPUs, was implemented. Our experimental results show that this implementation can execute the same source code efficiently on multi-core CPU platforms and GPUs, achieving an average speedup of 1.6 ~ 2.5x over previous implementations of MapReduce on eight commonly used applications.

Sponge

Graphics processing units (GPUs) provide a low cost platform for accelerating high performance computations. The introduction of new programming languages, such as CUDA and OpenCL, makes GPU programming attractive to a wide variety of programmers. However, programming GPUs is still a cumbersome task for two primary reasons: tedious performance optimizations and lack of portability. First, optimizing an algorithm for a specific GPU is a time-consuming task that requires a thorough understanding of both the algorithm and the underlying hardware. Unoptimized CUDA programs typically only achieve a small fraction of the peak GPU performance. Second, GPU code lacks efficient portability as code written for one GPU can be inefficient when executed on another. Moving code from one GPU to another while maintaining the desired performance is a non-trivial task often requiring significant modifications to account for the hardware differences. In this work, we propose Sponge, a compilation framework for GPUs using synchronous data flow streaming languages. Sponge is capable of performing a wide variety of optimizations to generate efficient code for graphics engines. Sponge alleviates the problems associated with current GPU programming methods by providing portability across different generations of GPUs and CPUs, and a better abstraction of the hardware details, such as the memory hierarchy and threading model. Using streaming, we provide a write-once software paradigm and rely on the compiler to automatically create optimized CUDA code for a wide variety of GPU targets. Sponge's compiler optimizations improve the performance of the baseline CUDA implementations by an average of 3.2x.

Sponge

Graphics processing units (GPUs) provide a low cost platform for accelerating high performance computations. The introduction of new programming languages, such as CUDA and OpenCL, makes GPU programming attractive to a wide variety of programmers. However, programming GPUs is still a cumbersome task for two primary reasons: tedious performance optimizations and lack of portability. First, optimizing an algorithm for a specific GPU is a time-consuming task that requires a thorough understanding of both the algorithm and the underlying hardware. Unoptimized CUDA programs typically only achieve a small fraction of the peak GPU performance. Second, GPU code lacks efficient portability as code written for one GPU can be inefficient when executed on another. Moving code from one GPU to another while maintaining the desired performance is a non-trivial task often requiring significant modifications to account for the hardware differences. In this work, we propose Sponge, a compilation framework for GPUs using synchronous data flow streaming languages. Sponge is capable of performing a wide variety of optimizations to generate efficient code for graphics engines. Sponge alleviates the problems associated with current GPU programming methods by providing portability across different generations of GPUs and CPUs, and a better abstraction of the hardware details, such as the memory hierarchy and threading model. Using streaming, we provide a write-once software paradigm and rely on the compiler to automatically create optimized CUDA code for a wide variety of GPU targets. Sponge's compiler optimizations improve the performance of the baseline CUDA implementations by an average of 3.2x.