Skip to main navigation Skip to search Skip to main content

Testing of execution of concurrent processes

Research output: Chapter in Book/Report/Conference proceedingConference paperResearchpeer-review

4 Citations (Scopus)

Abstract

Authors propose an algorithm for analysis of business processes to detect potentially incorrect results of concurrent processes execution. Our novel approach is to conclude necessary database isolation level from business process description. If traditional languages with loops and arithmetic operations (two-way counters) are used for business process descriptions, the problem of detecting incorrect execution of concurrent processes cannot be algorithmically solved. This paper introduces a simplified business processes description language CPL-1, a transaction mechanism and an algorithm that supports detection of incorrect results during the concurrent execution of business processes. Business processes are often run concurrently in real world tasks like billing systems, ticket distribution, hotel reservations, etc. Currently there are some popular solutions preventing incorrect execution of concurrent business processes by using built-in transaction mechanisms and/or resource reservations in database management systems (DBMS). The proposed solution is an alternative, which can be used when resource locking or DBMS transaction mechanisms cannot be applied.

Original languageEnglish
Title of host publicationDatabases and Information Systems - 14th International Baltic Conference, DB and IS 2020, Proceedings
EditorsTarmo Robal, Hele-Mai Haav, Jaan Penjam, Raimundas Matulevicius
Pages265-279
Volume1243 CCIS
DOIs
Publication statusPublished - 2020

Publication series

NameCommunications in Computer and Information Science
Volume1243 CCIS
ISSN (Print)1865-0929
ISSN (Electronic)1865-0937

Keywords

  • Concurrent transactions
  • Symbolic execution
  • Transaction execution state

Fingerprint

Dive into the research topics of 'Testing of execution of concurrent processes'. Together they form a unique fingerprint.

Cite this