A Block Structured Multimedia Data Prefetching 


Vol. 29,  No. 1, pp. 53-64, Jan.  2004


PDF
  Abstract

As to medium data which is involved in the form of streaming for a multimedia application, it characterizes that spatial locality occurs strongly but temporal locality appears even weaker. In this paper, with regard to dynamic prefetching, we suggest a method to make the most of memory reference regularities which typically innate by nature in the multimedia data with strong spatial locality but with weak temporal locality. Especially, the suggested method has a remarkable capability such that it can reduce prefeching errors substantially compared to existing prefetching methods for an application program which divides an array into small sub-blocks and, plus executes in the unit of sub-block. We carried out experiments to test the suggested method using various MediaBench benchmarks. From the results, we have confirmed that the occurrences of prefetching error decrease effectively than those of existing linear prefetching methods.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Cite this article

[IEEE Style]

S. Kim, B. Lee, S. Kim, "A Block Structured Multimedia Data Prefetching," The Journal of Korean Institute of Communications and Information Sciences, vol. 29, no. 1, pp. 53-64, 2004. DOI: .

[ACM Style]

Suk-Ju Kim, Byung-Kwon Lee, and Suk-Il Kim. 2004. A Block Structured Multimedia Data Prefetching. The Journal of Korean Institute of Communications and Information Sciences, 29, 1, (2004), 53-64. DOI: .

[KICS Style]

Suk-Ju Kim, Byung-Kwon Lee, Suk-Il Kim, "A Block Structured Multimedia Data Prefetching," The Journal of Korean Institute of Communications and Information Sciences, vol. 29, no. 1, pp. 53-64, 1. 2004.

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880