high performance computing on graphics processing units: hgpu.org

Field Programmable Gate Arrays (FPGAs) have been widely used for accelerating machine learning algorithms. However, the high design cost and time for implementing FPGA-based accelerators using traditional HDL-based design methodologies has discouraged users from designing FPGA-based accelerators. In recent years, a new CAD tool called Intel FPGA SDK for OpenCL (IFSO) allowed fast and efficient […]

This paper describes NCRF++, a toolkit for neural sequence labeling. NCRF++ is designed for quick implementation of different neural sequence labeling models with a CRF inference layer. It provides users with an inference for building the custom model structure through configuration file with flexible neural feature design and utilization. Built on PyTorch, the core operations […]

This paper presents the design and implementation of general matrix-matrix multiplication (GEMM) algorithm for the second generation Intel Xeon Phi processor codenamed Knights Landing (KNL). We illustrate several developing guidelines to achieve optimal performance with C programming language and the Advanced Vector Extensions (AVX-512) instruction set. Further, we present several environment variable issues associated with […]

Automatic text categorization presents many difficulties. Modern algorithms are getting better in extracting meaningful information from human language. However, they often significantly increase complexity of computations. This increased demand for computational capabilities can be facilitated by the usage of hardware accelerators like general purpose graphic cards. In this paper we present a full processing flow […]

Linear operators used in iterative methods like conjugate gradient have typically been implemented either as "matrix-driven" subroutines backed by explicit sparse or dense matrices, or as "matrix-free" subroutines that implement specific linear operations directly (e.g. FFTs). The matrix-driven approach is generally more portable because it can target widely available BLAS libraries, but it can be […]

Writing mixed-precision kernels allows to achieve higher throughput together with outputs whose precision remain within given limits. The recent introduction of native half-precision arithmetic capabilities in several GPUs, such as NVIDIA P100 and AMD Vega 10, contributes to make precision-tuning even more relevant as of late. However, it is not trivial to manually find which […]

We present a new method for Product Quantization (PQ) based approximated nearest neighbor search (ANN) in high dimensional spaces. Specifically, we first propose a quantization scheme for the codebook of coarse quantizer, product quantizer, and rotation matrix, to reduce the cost of accessing these codebooks. Our approach also combines a highly parallel k-selection method, which […]

Sparse matrix vector multiplication (SpMV) is one of the most common operations in scientific and high-performance applications, and is often responsible for the application performance bottleneck. While the sparse matrix representation has a significant impact on the resulting application performance, choosing the right representation typically relies on expert knowledge and trial and error. This paper […]

The integration of graphics processing units (GPUs) on high-end compute nodes has established a new accelerator-based heterogeneous computing model, which now permeates high performance computing. The same paradigm nevertheless has limited adoption in cloud computing or other large-scale distributed computing paradigms. Heterogeneous computing with GPUs can benefit the Cloud by reducing operational costs and improving […]

Convolutional neural networks (CNNs) are widely used in many computer vision applications. Previous FPGA implementations of CNNs are mainly based on the conventional convolutional algorithm. However, the high arithmetic complexity of conventional convolution algorithm for CNNs restricts the performance of accelerators and significantly increases the challenges of design. It has been proved that the Winograd […]

We present a GPU-based Locality Sensitive Hashing (LSH) algorithm to speed up beam search for sequence models. We utilize the winner-take-all (WTA) hash, which is based on relative ranking order of hidden dimensions and thus resilient to perturbations in numerical values. Our algorithm is designed by fully considering the underling architecture of CUDA-enabled GPUs (Algorithm/Architecture […]

This paper presents a novel generative model to synthesize fluid simulations from a set of reduced parameters. A convolutional neural network is trained on a collection of discrete, parameterizable fluid simulation velocity fields. Due to the capability of deep learning architectures to learn representative features of the data, our generative model is able to accurately […]