A Multi-Node Co-Allocation Parallel Downloading Algorithm

Qing Xiang Wu; Jian Mei Shuai; Xiao Shuang Lu

doi:10.4028/www.scientific.net/AMM.58-60.1590

Paper Titles

Research of the Wireless Sensor Networks Routing Protocol Based on CEJ Ant Colony Algorithm
p.1566

Wyner-Ziv Coding over Broadcast Channel with Noncausal Channel Side Information only Known at Encoder
p.1572

Dynamic Analysis and Control Experiment of a Novel Mobile Robot
p.1577

Research on Converged Access Network of WDM-PON and WiMAX
p.1583

A Multi-Node Co-Allocation Parallel Downloading Algorithm
p.1590

Accelerating the 3GPP LTE System Level Simulation with NVidia CUDA
p.1596

Adaptive Fuzzy PID Arithmetic Applied in Water Tank Temperature Control
p.1602

Multibody Dynamic Analysis of Automobile Wiper Based on Virtual Prototype
p.1608

The Management of Play Buffer and Arithmetic of Nodes Selecting in P2P Streaming Media
p.1614

HomeApplied Mechanics and MaterialsApplied Mechanics and Materials Vols. 58-60A Multi-Node Co-Allocation Parallel Downloading...

A Multi-Node Co-Allocation Parallel Downloading Algorithm

Abstract:

In distributed data storage, a particular dataset can reside at multiple locations in order to get high availability .Thus, the dataset can be downloaded in parallel from multiple nodes.Throughput between server and client changes dynamically, so the downloading speed can vary unpredictably. A dynamic parallel downloading algorithm based on measurement of bandwidth and bandwidth prediction is produced in this paper and server caching is adopted in order to improve downloading speed. The algorithm dynamically adjusts downloading of the last block to make parallel downloading from multiple servers end almost simultaneously. With this approach, the download time is reduced and the robustness of the downloading system is improved. Besides, the algorithm not only avoids complicated server selecting mechanism, but also improves load balance of the servers.

You might also be interested in these eBooks

View Preview

Info:

Periodical:

Applied Mechanics and Materials (Volumes 58-60)

Pages:

1590-1595

DOI:

https://doi.org/10.4028/www.scientific.net/AMM.58-60.1590

Citation:

Cite this paper

Online since:

June 2011

Authors:

Qing Xiang Wu, Jian Mei Shuai, Xiao Shuang Lu

Keywords:

Dynamic Scheduling Mechanism, Multi-Node Co-Allocation, Parallel Downloading, Server Caching

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] S. Ghemawat, Howard Gobioff and S. T. Leung, The Google File System", SOSP, 03, Bolton Landing, New York, USA, October 19–22, (2003).

Google Scholar

[2] HDFS, http: /hadoop. apache. org/hdfs.

Google Scholar

[3] G. DeCandia, D. Hastorun, M. Jampani and et al., Dynamo: Amazon's Highly Available Key-value Store", SOSP, 07, Stevenson, Washington, USA, October 14–17, (2007).

DOI: 10.1145/1294261.1294281

Google Scholar

[4] S. Vazhkudai, Enabling the Co-Allocation of Grid Data Transfers", GRID, 03.

Google Scholar

[5] R. S. Chang, M. H. Guo and H. C. Lin, A multiple parallel download scheme with server throughput and client bandwidth considerations for data grids, Future Generation Computer Systems 24 (2008) 798–805.

DOI: 10.1016/j.future.2008.04.006

Google Scholar

[6] B. Lertnitiwong and L. Yang, Performance Evaluation of the Multi-Source File Transfer Protocol in Peer-to-Peer Network", SpringSim , 09.

Google Scholar

[7] C. T. Yang, M. F. Yang and W. C. Chiang, Enhancement of anticipative recursively adjusting mechanism for redundant parallel ﬁle transfer in data grids, Journal of Network and Computer Applications 32 , p.834–845, (2009).

DOI: 10.1016/j.jnca.2009.02.002

Google Scholar

[8] Memcached, http: /memcached. org.

Google Scholar

[9] Iperf, http: /sourceforge. net/projects/iperf.

Google Scholar

[10] Ethtool, http: /sourceforge. net/projects/gkernel.

Google Scholar