An Transfer Latency Optimized Solution in GPU-Accelerated De-Duplication

Rui Zhu; Chang Nian Chen; Lei Hua Qin

doi:10.4028/www.scientific.net/AMM.336-338.2059

Paper Titles

Design of a New Intelligence Support System
p.2040

The Design and Realization of Lift Balance Coefficient Calculation Software Based on Android Smartphone
p.2044

Research of Cloud Computing Information Management Mode for Oil Enterprise
p.2048

Research on Graphic Displaying Algorithm for Electron Beam Lithography
p.2054

An Transfer Latency Optimized Solution in GPU-Accelerated De-Duplication
p.2059

Test Case Generation of Web Service Composition: An Approach Based on the Color Petri Net
p.2063

The Construction and Design of Personalized Course Website
p.2071

The Study on the Implementation of AOP in .NET Platform
p.2077

Using the ORACLE Database Management for 500kV Transmission Line Fault Features
p.2083

HomeApplied Mechanics and MaterialsApplied Mechanics and Materials Vols. 336-338An Transfer Latency Optimized Solution in...

An Transfer Latency Optimized Solution in GPU-Accelerated De-Duplication

Abstract:

Recently, GPU has been introduced as an important tool in general purpose programming due to its powerful computing capacity. In data de-duplication systems, GPU has been used to accelerate the chunking and hashing algorithms. However, the data transfer latency between the memories of CPU to GPU is one of the main challenges in GPU accelerated de-duplication. To alleviate this challenge, our solution strives to reduce the data transfer time between host and GPU memory on parallelized content-defined chunking and hashing algorithm. In our experiment, it has shown 15%~20% performance improvements over already accelerated baseline GPU implementation in data de-duplication.

You might also be interested in these eBooks

Industrial Instrumentation and Control Systems II

View Preview

Info:

Periodical:

Applied Mechanics and Materials (Volumes 336-338)

Pages:

2059-2062

DOI:

https://doi.org/10.4028/www.scientific.net/AMM.336-338.2059

Citation:

Cite this paper

Online since:

July 2013

Authors:

Rui Zhu*, Chang Nian Chen, Lei Hua Qin

Keywords:

Content-Defined Chunking, Data De-Duplication, GPGPU

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

* - Corresponding Author

References

[1] Moya V. and Gonzalez, C. and Roca, J. and Fernandez, A. and Espasa, R. Shader performance analysis on a modern GPU architecture, In IEEE/ACM International Symposium on Microarchitecture, MICRO-38, pp.10-22, (2005).

DOI: 10.1109/micro.2005.30

Google Scholar

[2] INTRODUCING CUDA 5. http: /www. nvidia. com/object/cuda_home_new. html.

Google Scholar

[3] Eshghi, K. and Lillibridge, M. and Wilcock, L. and Belrose, G. and Hawkes, R. Jumbo Store: Providing Efficient Incremental Upload and Versioning for a Utility Rendering Service, In Proceedings of the 5th USENIX conference on File and Storage Technologies, pp.22-33, (2007).

Google Scholar

[4] GPGPU community website. http: /www. gpgpu. org.

Google Scholar

[5] Al-Kiswany Samer, Gharaibeh Abdullah, Santos-Neto Elizu, Yuan George, Ripeanu Matei. StoreGPU: Exploiting Graphics Processing Units to Accelerate Distributed Storage Systems, In Proceedings of the 17th international symposium on High Performance Distributed Computing, pp.165-174, (2008).

DOI: 10.1145/1383422.1383443

Google Scholar

[6] Udi Manber. Finding Similar Files in a Large File System, In proceedings of the USENIX Winter 1994 Technical Conference, pp.17-21, (1994).

Google Scholar

[7] J. Ziv and A. Lempel. A universal algorithm for sequential data compression, IEEE Transactions on Information Theory, Vol. 23, No. 3, p.337~343, (1997).

DOI: 10.1109/tit.1977.1055714

Google Scholar