Improving the Performance of Content-Locality Based Storage Management

Article Preview

Abstract:

The wide spread of cloud services has increased the demand for high performance storages. There have been various research efforts to improve performance by combining HDDs and SDDs in a hybrid form because the price per unit capacity of SSDs is higher than that of HDDs. A method that exploits the characteristic of content locality has been proposed and it reduces sizes of data by using delta encoding. However the method has limitations if data has low content locality. This paper modifies the method to overcome such limitations. The proposed method is able to reduce the overhead of delta encoding by skipping the encoding if data are found to have low locality. Instead, it utilizes caching to improve the IO performance of such data. Experiment results show that the performance of the proposed method is at least 20% higher than the original scheme.

You might also be interested in these eBooks

Info:

Periodical:

Pages:

370-374

Citation:

Online since:

July 2013

Export:

Price:

Permissions CCC:

Permissions PLS:

Сopyright:

© 2013 Trans Tech Publications Ltd. All Rights Reserved

Share:

Citation:

[1] L. Chang: Hybrid solid-state disks: Combining heterogeneous NAND flash in large SSDs, in Proc. of Design Automation Conference, pp.428-433, Mar. (2008)

DOI: 10.1109/aspdac.2008.4483988

Google Scholar

[2] L. Chang and T. Kuo: Efficient Management for Large-Scale Flash-Memory Storage Systems with Resource Conservation, ACM Transactions on Storage, Vol.1, Issue 4, (2005)

DOI: 10.1145/1111609.1111610

Google Scholar

[3] E. Seppanen, M. OKeefe, and D. Jilja: High Performance Solid State Storage under Linux, in Proc. of the 26th IEEE Symposium on MSST, pp.1-12, May 2010.

DOI: 10.1109/msst.2010.5496976

Google Scholar

[4] M. Dunn, and A. Reddy: A new I/O scheduler for solid state devices. Texas A&M University ECE Technical Report TAMU-ECE-2009-02, 2009.

Google Scholar

[5] J. Kim and J. Oh, et al.: Disk schedulers for solid state drivers, in Proc. of the 7th ACM international conf. on Embedded software, pp.295-304, 2009.

Google Scholar

[6] S. Kang, H. Park and C. Yoo: Performance Enhancement of I/O Scheduler for Solid State Devices, in Proc. of 2011 IEEE ICCE, pp.31-32, Jan., 2011.

DOI: 10.1109/icce.2011.5722605

Google Scholar

[7] J. Ren: I-CASH, Intelligently Coupled Array of SSD and HDD, in Proc. of 17th International Symposium on High Performance Computer Architecture, pp.278-289, Feb. 2011.

DOI: 10.1109/hpca.2011.5749736

Google Scholar

[8] http://www.iozone.org

Google Scholar

[9] http:// www.tpc.org/tpcc/

Google Scholar