high performance computing on graphics processing units: hgpu.org

Python is rapidly becoming the lingua franca of machine learning and scientific computing. With the broad use of frameworks such as Numpy, SciPy, and TensorFlow, scientific computing and machine learning are seeing a productivity boost on systems without a requisite loss in performance. While high-performance libraries often provide adequate performance within a node, distributed computing […]

Strong demands for efficient deployment of Deep Learning (DL) applications prompt the rapid development of a rich DL ecosystem. To keep up with its fast advancement, it is crucial for DL frameworks to efficiently integrate a variety of optimized libraries and runtimes as their backends and generate the fastest possible executable by using them properly. […]

Newer use cases of GPU (Graphics Processing Unit) computing, e.g., graph analytics, look less like traditional bulksynchronous GPU programs. To cater to the needs of emerging applications with semantically richer and finer grain sharing patterns, GPU vendors have been introducing advanced programming features, e.g., scoped synchronization and independent thread scheduling. While these features can speed […]

FPGAs are an interesting platform for the implementation of network-attached accelerators, either in the form of smart network interface cards or as In-Network Processing accelerators. Both application scenarios require a high-throughput hardware network stack. In this work, we integrate such a stack into the open-source TaPaSCo framework and implement a library of easy-to-use design primitives […]

Physics-based simulation has been actively employed in generating offline visual effects in the film and animation industry. However, the computations required for high-quality scenarios are generally immense, deterring its adoption in real-time applications, e.g., virtual production, avatar live-streaming, and cloud gaming. We summarize the principles that can accelerate the computation pipeline on single-GPU and multi-GPU […]

This paper presents our work toward correct and efficient automatic differentiation of OpenMP parallel worksharing loops in forward and reverse mode. Automatic differentiation is a method to obtain gradients of numerical programs, which are crucial in optimization, uncertainty quantification, and machine learning. The computational cost to compute gradients is a common bottleneck in practice. For […]

The massive parallelism provided by general-purpose GPUs (GPGPUs) possessing numerous compute threads in their streaming multiprocessors (SMs) and enormous memory bandwidths have made them the de-facto accelerator of choice in many scientific domains. To support the complex memory access patterns of applications, GPGPUs have a multi-level memory hierarchy consisting of a huge register file and […]

The Transformer architecture has improved the performance of deep learning models in domains such as Computer Vision and Natural Language Processing. Together with better performance come larger model sizes. This imposes challenges to the memory wall of the current accelerator hardware such as GPU. It is never ideal to train large models such as Vision […]

Modern graphics computing units (GPUs) are designed and optimized to perform highly parallel numerical calculations. This parallelism has enabled (and promises) significant advantages, both in terms of energy performance and calculation. In this document, we take stock of the different applications of mixed precision. We recall the standards currently used in the overwhelming majority of […]

This document describes version 0.10 of torchaudio: building blocks for machine learning applications in the audio and speech processing domain. The objective of torchaudio is to accelerate the development and deployment of machine learning applications for researchers and engineers by providing off-the-shelf building blocks. The building blocks are designed to be GPU-compatible, automatically differentiable, and […]

Today’s auto-tuners (e.g., AutoTVM, Ansor) generate efficient tensor programs by navigating a large search space to identify effective implementations, but they do so with opaque hardware details. Thus, their performance could fall behind that of hardware-native libraries (e.g., cuBLAS, cuDNN), which are hand-optimized by device vendors to extract high performance. On the other hand, these […]

There is often variation in the shape and size of input data used for deep learning. In many cases, such data can be represented using tensors with non-uniform shapes, or ragged tensors. Due to limited and non-portable support for efficient execution on ragged tensors, current deep learning frameworks generally use techniques such as padding and […]