Information theoretical complexity metrics for concurrent programs병행 프로그램을 위한 정보이론적 복잡도 척도

Software complexity is acutely recognized as the major reason for rapidly increasing software development cost. To deal with this problem effectively, software complexity metrics have been perceived as a critical element of software development. With the increased use of concurrent and distributed computing in wide range of applications, a need exists for techniques which can be used to aid in development of correct and reliable distributed software. Controlling, or at least recognizing, complexity of such software is an important part of the development and maintenance process. But, despite numerous metrics to measure the complexity of sequential programs, software complexity metrics are almost nonexistent for concurrent programs. In this dissertation, we propose a model and metrics which quantize the complexity of concurrent programs. As primary factors that complicate the processes of understanding and developing of concurrent programs, information content, the amount of intertasking activity, nondeterminism and accessibility are identified. The primary notion of program complexity germane to this dissertation is a notion of the quantified amount of information needed to resolve any uncertainty manifested through the program. Hence, the proposed metrics are based on the entropy function of communication theory which is well-justified measures of uncertainty. An execution of a concurrent program can be interpreted as an experiment that selects, from all possible ways that a program allows, the single correct configuration for that execution. The information content is used to measure the uncertainty contained in this experiment. The metric for intertasking measures the amount of information caused through task coordination. And the metric for nondeterminism can be used to measure the uncertainty which may be occurred by nondeterministic features in a concurrent program. The information content, the amount of intertasking activity and the nondeterminism cont...
Advisors
Kwon, Yong-Rae(권용래)researcher
Publisher
한국과학기술원
Issue Date
1995
Identifier
99170/325007 / 000925375
Language
eng
Description

학위논문(박사) - 한국과학기술원 : 전산학과, 1995.2, [ vii, 118 p. ]

Keywords

프로그램 복잡도; 병렬 프로그램

URI
http://hdl.handle.net/10203/33036
Link
http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=99170&flag=t
Appears in Collection
CS-Theses_Ph.D.(박사논문)
Files in This Item
There are no files associated with this item.
  • Hit : 55
  • Download : 0
  • Cited 0 times in thomson ci

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0