Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Search
Locality analysis: a nonillion time window problem
Brock J., Luo H., Ding C. ACM SIGMETRICS Performance Evaluation Review41 (4):102-105,2014.Type:Article
Date Reviewed: Jul 11 2014

Brock et al., in this paper, revisit the problem of instruction/data locality of references (temporal and spatial), which plays a key role in the latency of instructions being executed by the central processing unit (CPU) from the memory. The general principle of locality suggests that the closer data are to the CPU, the faster they are processed. The paper positions locality as a big data problem by introducing the notion of the footprint of data accessed effectively by the CPU within a predefined window of execution. It is shown that, as the length of the execution increases, the number of CPU cycles required increases exponentially.

This contribution is not news and is well documented in the literature on computer architecture. The theory on caching and locality (in computer architecture) is slowly but surely being employed in software development in order to resolve some latency-related challenges. The notion of caching currently employed by web browsers is a typical example. It is thus no surprise that we, in the near future, will see more developers mimicking most of the techniques employed at the hardware level to overcome some software challenges for performance enhancement. The development of cache-optimized algorithms has been particularly attractive to researchers working on high-performance pattern matching, for example. Furthermore, current technological trends reveal that “in-memory computing” is increasingly gaining momentum and appears to be the most promising solution to the challenge of big data (both structured and unstructured). Of course, there is no harm in combining in-memory technology with cache-optimized algorithms for further performance enhancement.

In conclusion, it is difficult to pinpoint the exact contribution of this paper in the following sense: 1) locality analysis is indeed a nonillion-time window problem and it is not new; 2) the use of theories from computer architecture to resolve software development problems is also not new; 3) problems with web-search terms, website views, product sales, and others may indeed be resolved through locality theory, but we should bear in mind that there are already solutions to those challenges and a cross-comparison exercise would be required to practically qualify the merit of locality theory. The most plausible contribution of the paper would have been to undertake an empirical study of a given real-life problem (for example, web search term) with a solution relying on locality analysis to demonstrate the relevance and applicability of the theory beyond the chip.

Reviewer:  Ernest Ketcha Ngassam Review #: CR142499 (1410-0869)
Bookmark and Share
  Reviewer Selected
 
 
Metrics (D.2.8 )
 
Would you recommend this review?
yes
no
Other reviews under "Metrics": Date
A comparison of time domains for software reliability models
Musa J., Okumoto K. Journal of Systems and Software 4(4): 277-287, 1984. Type: Article
May 1 1985
On software equations
Král J. Information Processing Letters 19(4): 191-196, 1984. Type: Article
Jun 1 1985
Software metrics: establishing a company-wide program
Grady R., Caswell D., Prentice-Hall, Inc., Upper Saddle River, NJ, 1987. Type: Book (9789780138218447)
Apr 1 1988
more...

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2024 ThinkLoud®
Terms of Use
| Privacy Policy