Fast parallel skew and prefix‐doubling suffix array construction on the GPU

Wang, Leyuan; Baxter, Sean; Owens, John D.

doi:10.1002/cpe.3867

Cited by 13 publications

(4 citation statements)

References 32 publications

(80 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We used segmented sorting to optimize the argsort for CNN models as illustrated in Figure 2. Note that although our implementation uses similar ideas as in previous segmented sort work [16] and merge sort work [38], it is not limited to Nvidia GPUs but uses the unified program to run efficiently on different integrated GPUs with different architectures.…”

Section: Vision-specific Operatorsmentioning

confidence: 99%

A Unified Optimization Approach for CNN Model Inference on Integrated GPUs

Wang

Chen

Liu

et al. 2019

Proceedings of the 48th International Conference on Parallel Processing

Self Cite

View full text Add to dashboard Cite

Modern deep learning applications urge to push the model inference taking place at the edge devices for multiple reasons such as achieving shorter latency, relieving the burden of the network connecting to the cloud, and protecting user privacy. The Convolutional Neural Network (CNN ) is one of the most widely used model family in the applications. Given the high computational complexity of the CNN models, it is favorable to execute them on the integrated GPUs at the edge devices, which are ubiquitous and have more power and better energy efficiency than the accompanying CPUs. However, programming on integrated GPUs efficiently is challenging due to the variety of their architectures and programming interfaces. This paper proposes an end-to-end solution to execute CNN model inference on the integrated GPUs at the edge, which uses a unified IR to represent and optimize vision-specific operators on integrated GPUs from multiple vendors, as well as leverages machine learning-based scheduling search schemes to optimize computationally-intensive operators like convolution. Our solution even provides a fallback mechanism for operators not suitable or convenient to run on GPUs. The evaluation results suggest that compared to state-of-the-art solutions backed up by the vendorprovided high-performance libraries on Intel Graphics, ARM Mali GPU, and Nvidia integrated Maxwell GPU, our solution achieves similar, or even better (up to 1.62×), performance on a number of popular image classification and object detection models. In addition, our solution has a wider model coverage and is more flexible to embrace new models. Our solution has been adopted in production services in AWS and is open-sourced.

show abstract

Section: Vision-specific Operatorsmentioning

confidence: 99%

A Unified Optimization Approach for CNN Model Inference on Integrated GPUs

Wang

Chen

Liu

et al. 2019

Proceedings of the 48th International Conference on Parallel Processing

Self Cite

View full text Add to dashboard Cite

show abstract

“…We compare our GPU SACA with three shared memory parallel SACAs. We could not compare our algorithm with the most recent GPU-algorithm by Wang et al [47], we could only run it successfully for inputs smaller than 100 KiB for our text collection. (To test their code, they use somehow meaningless random input texts, which we could get to work in our test environment for sizes up to 170 MiB.…”

Section: Evaluation Of Parallel Suffix Array Construction Algorithmsmentioning

confidence: 99%

“…It is generally accepted that the Divsufsort [12,33] is the fastest SACA-despite it having a superlinear running time. Different models of computation have also been considered for this problem: external memory, e. g., [5,9,18,19,20,38], shared memory, e. g., [20,20,25], distributed memory, e. g., [1,6,13,14,20,32,36], and GPGPU, e. g., [10,41,46,47].…”

Section: Introductionmentioning

confidence: 99%

SACABench: Benchmarking Suffix Array Construction

Bahne

Bertram

Böcker

et al. 2019

String Processing and Information Retrieval

View full text Add to dashboard Cite

We present a practical comparison of suffix array construction algorithms on modern hardware. The benchmark is conducted using our new benchmark framework SACABench, which allows for an easy deployment of publicly available implementations, simple plotting of the results, and straight forward support to include new construction algorithms. We use the framework to develop a construction algorithm running on the GPU that is competitive with the fastest parallel algorithm in our test environment.

show abstract

“…The paper Fast parallel skew and prefix‐doubling suffix array construction on the GPU , authored by Leyuan Wang, Sean Baxter and John D. Owens , proposes a hybrid GPU implementation of known algorithms for constructing suffix arrays of a string that fits the given GPU architecture best. One highlight pointed out in the reviews is a highly efficient segmented sorting primitive, which is also valuable as independent result.…”

mentioning

confidence: 99%

Special issue: Euro‐Par 2015

Lengauer

Bougé

Träff

2016

Concurrency and Computation

View full text Add to dashboard Cite

This special issue of Concurrency and Computation: Practice and Experience contains revised and extended versions of selected papers presented at the conference Euro-Par 2015.Euro-Par-the European Conference on Parallel Computing-is an annual series of international conferences dedicated to the promotion and advancement of all aspects of parallel and distributed computing. Euro-Par covers a wide spectrum of topics from algorithms and theory to software technology and hardware-related issues, with application areas ranging from scientific to mobile and cloud computing. The major part of the Euro-Par audience consists of researchers in academic institutions, government laboratories and industrial organisations.Euro-Par 2015, the 21st conference in the Euro-Par series, was held in Vienna, Austria. It was organised by the Research Group for Parallel Computing of the Vienna University of Technology (TU Wien).Thirteen broad topics were defined and advertised, covering a large variety of aspects of parallel and distributed computing. The call for papers attracted a total of 190 submissions. The submitted papers were reviewed at least three and, in most cases, four or even more times (four reviews on average). A total of 51 papers were finally accepted for publication. This makes a global acceptance rate of 27 %. The authors of accepted papers came from 21 countries, with the four main contributing countries-the United States, France, Spain and Germany-accounting for a bit more than half of them.Based on the results of the reviews and a majority opinion of the respective topic programme committees, a number of papers were recommended for this special issue. The authors were contacted at the conference and invited to submit revised and extended versions of their papers. These new versions were reviewed independently by three reviewers; two had previously reviewed the conference version, the third had not. Eventually, four papers were accepted for publication. This year, two Euro-Par topics are represented-both covering methods of programming modern computer architectures.Topic 13 on Accelerator Computing is represented with three papers. The paper Performance optimization of sparse matrix-vector multiplication for multi-component PDE-based applications using GPUs, authored by Ahmad Abdelfattah, Hatem Ltaief, David Keyes and Jack Dongarra [1], describes the implementation of a single-GPU and multi-GPU kernel for block-sparse matrix-vector multiplication, a problem that appears in the discretisation of partial differential equations with many dependent variables. The performance of the kernel is measured on a subset of the Florida Sparse Matrix Collection. Especially noted by the reviewers was the uniform interface that applies to a wide range of problem sizes via tunable parameters. This makes it perform efficiently on a wide range of GPU architectures running CUDA.The paper Fast parallel skew and prefix-doubling suffix array construction on the GPU, authored by Leyuan Wang, Sean Baxter and John D. Owens [2], proposes a h...

show abstract

Fast parallel skew and prefix‐doubling suffix array construction on the GPU

Cited by 13 publications

References 32 publications

A Unified Optimization Approach for CNN Model Inference on Integrated GPUs

A Unified Optimization Approach for CNN Model Inference on Integrated GPUs

SACABench: Benchmarking Suffix Array Construction

Special issue: Euro‐Par 2015

Contact Info

Product

Resources

About