izpis_h1_title_alt

Vzporedno programiranje na osnovi skupnega prostora terk v OCamlu
ID Gazvoda de Reggi, Marin (Author), ID Slivnik, Boštjan (Mentor) More about this mentor... This link opens in a new window

.pdfPDF - Presentation file, Download (343,28 KB)
MD5: 4E007CCC804F4720D144F0052392A6F2

Abstract
V diplomskem delu je obravnavana implementacija modela Linda za vzporedno programiranje v funkcijskem programskem jeziku OCaml. Model Linda omogoča elegantno koordinacijo vzporednih procesov preko skupnega prostora terk. Naš pristop vključuje razvoj knjižnice, ki implementira osnovne operacije modela Linda, pri čemer smo izkoristili nove zmožnosti za vzporedno programiranje v OCamlu 5.0. Ključni prispevek je učinkovita implementacija prostora terk z uporabo drevesne podatkovne strukture in večnivojskega zaklepanja, kar nudi boljšo skalabilnost in višjo stopnjo sočasnosti v primerjavi z naivno implementacijo. Eksperimentalni rezultati kažejo, da naša implementacija dosega dobro učinkovitost in skalabilnost, zlasti pri večjem številu vzporednih procesov in kompleksnejših vzorcih uporabe.

Language:Slovenian
Keywords:model Linda, OCaml, vzporednost, prostor terk
Work type:Bachelor thesis/paper
Typology:2.11 - Undergraduate Thesis
Organization:FRI - Faculty of Computer and Information Science
Year:2024
PID:20.500.12556/RUL-160943 This link opens in a new window
COBISS.SI-ID:209716995 This link opens in a new window
Publication date in RUL:05.09.2024
Views:205
Downloads:38
Metadata:XML DC-XML DC-RDF
:
Copy citation
Share:Bookmark and Share

Secondary language

Language:English
Title:Tuple Space-Based Parallel Programming in OCaml
Abstract:
This thesis presents an implementation of the Linda model for parallel programming in the functional language OCaml. Linda offers an elegant approach to coordinating parallel processes through a shared tuple space. We have developed a library that implements Linda's core operations, leveraging the enhanced parallel programming capabilities introduced in OCaml 5.0. The primary contribution of this work is an efficient tuple space implementation utilizing a tree-based data structure and multi-level locking mechanism. This approach significantly improves scalability and concurrency compared to naive implementations. Our experimental results demonstrate the implementation's effectiveness and scalability, particularly when dealing with a large number of parallel processes and complex usage scenarios.

Keywords:Linda model, OCaml, parallelism, tuple space

Similar documents

Similar works from RUL:
Similar works from other Slovenian collections:

Back