Please use this identifier to cite or link to this item: https://www.um.edu.mt/library/oar/handle/123456789/77683
Title: Separation-based reasoning for deterministic message-passing concurrent programs
Authors: Borda, Aimee (2013)
Keywords: Parallel processing (Electronic computers)
Sorting (Electronic computers)
List processing (Electronic computers)
Issue Date: 2013
Citation: Borda, A. (2013). Separation-based reasoning for deterministic message-passing concurrent programs (Master’s dissertation).
Abstract: The correctness of concurrent programs is hard to reason about because of the multiple interleaving and interactions between the independently executing program entities (e.g., threads, processes etc.) that need to be considered. Yet, for a substantial subset of concurrent programs that are required to be deterministic, more scalable and modular proof systems have been proposed, e.g., [O'H07, PM06, VP07, DFPV09, FRS11]. In these proof systems only a subset of interleaving/interactions are considered and the analysis is decomposed so that the correctness of the global program is derived from the analysis of its sub-programs, i.e., local reasoning. In this work we study the development of a local reasoning technique, based on resource separation, to reason about the correctness of message-passing concurrent programs. In particular we focus on reuse patterns traditionally used in this setting. We demonstrate the feasibility of our logical framework through the analysis of a resource-efficient implementation of Sorting Networks (SNs) [Bat68, Knu98]. A set of deterministic algorithms for sorting list of values concurrently, which are also used to give deeper insights into problems such as loop parallelisation.
Description: M.SC.COMP.SCI.&ARTIFICIAL INTELLIGENCE
URI: https://www.um.edu.mt/library/oar/handle/123456789/77683
Appears in Collections:Dissertations - FacICT - 2013
Dissertations - FacICTAI - 2002-2014

Files in This Item:
File Description SizeFormat 
M.SC.COMP.SCI._ARTIFICIAL INTELLIGENCE_Borda_Aimee_2013.pdf
  Restricted Access
6.92 MBAdobe PDFView/Open Request a copy


Items in OAR@UM are protected by copyright, with all rights reserved, unless otherwise indicated.