Singular value decomposition on GPU using CUDA

Lahabar, Sheetal; Narayanan, P. J.

doi:10.1109/ipdps.2009.5161058

Cited by 104 publications

(53 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In LSA the dimension of the vector space is reduced by singular value decomposition (SVD), which is a computationally demanding operation. By a bidiagonalization method, a speedup up to 8x has been achieved for large matrices [26]. The baseline CPU implementation used Basic Linear Algebra Subprograms (BLAS) calls of the highly optimized Intel Math Kernel Library on an Intel Dual Core 2.66GHz PC, whereas the GPU version relied on Compute Unified BLAS (CUBLAS) running on an NVIDIA GTX 280.…”

Section: Text Mining On Gpusmentioning

confidence: 99%

Accelerating text mining workloads in a MapReduce-based distributed GPU environment

Wittek

Daranyi

2013

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

Scientific computations have been using GPU-enabled computers successfully, often relying on distributed nodes to overcome the limitations of device memory. Only a handful of text mining applications benefit from such infrastructure. Since the initial steps of text mining are typically data-intensive, and the ease of deployment of algorithms is an important factor in developing advanced applications, we introduce a flexible, distributed, MapReducebased text mining workflow that performs I/O-bound operations on CPUs with industry-standard tools and then runs compute-bound operations on GPUs which are optimized to ensure coalesced memory access and effective use of shared memory. We have performed extensive tests of our algorithms on a cluster of eight nodes with two NVidia Tesla M2050 attached to each, and we achieve considerable speedups for random projection and self-organizing maps.

show abstract

Section: Text Mining On Gpusmentioning

confidence: 99%

Accelerating text mining workloads in a MapReduce-based distributed GPU environment

Wittek

Daranyi

2013

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

show abstract

“…Many of these tools are common but demand effort to be efficiently implemented into GPU formalism. Thankfully, they are progressively ported to CUDA, like the SVD implementation of Lahabar and Narayanan [15]. However, their integration into an existent framework is not always straightforward as special data structures are often required.…”

Section: Accuracy and Robustnessmentioning

confidence: 99%

A GPU framework for parallel segmentation of volumetric images using discrete deformable models

et al. 2010

View full text Add to dashboard Cite

Despite the ability of current GPU processors to treat heavy parallel computation tasks, its use for solving medical image segmentation problems is still not fully exploited and remains challenging. A lot of difficulties may arise related to, for example, the different image modalities, noise and artifacts of source images, or the shape and appearance variability of the structures to segment. Motivated by practical problems of image segmentation in the medical field, we present in this paper a GPU framework based on explicit discrete deformable models, implemented over the NVidia CUDA architecture, aimed for the segmentation of volumetric images. The framework supports the segmentation in parallel of different volumetric structures as well as interaction during the segmentation process and real-time visualization of the intermediate results. Promising results in terms of accuracy and speed on a real segmentation experiment have demonstrated the usability of the system.

show abstract

“…GPU-based QR decomposition was also studied by [8] using blocked Houserholder reflections. Recently, Lahabar et al [9] presented a GPUbased SVD algorithm built upon the Golub-Reinsch method. They achieve up to 8× speedup over an Intel MKL implementation running on dual core CPU.…”

Section: Introductionmentioning

confidence: 99%

A GPU-Based Approximate SVD Algorithm

Foster

Mahadevan

Wang

2012

Parallel Processing and Applied Mathematics

View full text Add to dashboard Cite

Abstract. Approximation of matrices using the Singular Value Decomposition (SVD) plays a central role in many science and engineering applications. However, the computation cost of an exact SVD is prohibitively high for very large matrices. In this paper, we describe a GPU-based approximate SVD algorithm for large matrices. Our method is based on the QUIC-SVD introduced by [6], which exploits a tree-based structure to efficiently discover a subset of rows that spans the matrix space. We describe how to map QUIC-SVD onto the GPU, and improve its speed and stability using a blocked Gram-Schmidt orthogonalization method. Results show that our GPU algorithm achieves 6∼7 times speedup over an optimized CPU version of QUIC-SVD, which itself is orders of magnitude faster than exact SVD methods. Using a simple matrix partitioning scheme, we have extended our algorithm to out-of-core computation, suitable for very large matrices that exceed the main memory size.

show abstract

Singular value decomposition on GPU using CUDA

Abstract: Abstract

Cited by 104 publications

References 16 publications

Accelerating text mining workloads in a MapReduce-based distributed GPU environment

Accelerating text mining workloads in a MapReduce-based distributed GPU environment

A GPU framework for parallel segmentation of volumetric images using discrete deformable models

A GPU-Based Approximate SVD Algorithm

Contact Info

Product

Resources

About