Architecture of software system for hierarchical fuzzy inference

A software architecture for hierarchical fuzzy logic hierarchy based on fuzzy logic algorithmic models and separate parallel computing technologies such as MPI and CUDA has been developed, as well as key architectural design solutions to provide the flexibility and performance of a parallel fuzzy sy...

Full description

Saved in:
Bibliographic Details
Published in:PROBLEMS IN PROGRAMMING
Date:2018
Issue:2-3
Pages:99-108
ISSN:1727-4907
Author Affiliations:
  • S.V. Yershov — Glushkov Institute of Cybernetics NAS of Ukraine
  • R.М.. Ponomarenko — Glushkov Institute of Cybernetics NAS of Ukraine
Keywords:шаблони паралельного проектування, нечітке логічне виведення, архітектура програмної системи, програмна архітектура, технологія паралельних обчислень, паралельна програма, логічне виведення, неоднорідні паралельні обчислювальні системи, розподілені системи та паралельне програмування, паралельні алгоритми
Main Authors: Yershov, S.V., Ponomarenko, R.М..
Format: Article
Language:Ukrainian
Published: PROBLEMS IN PROGRAMMING 2018
Subjects:
Online Access:https://pp.isofts.kiev.ua/index.php/ojs1/article/view/271
Tags: Add Tag
No Tags, Be the first to tag this record!
Journal Title:Problems in programming
Download file: Pdf

Institution

Problems in programming
Description
Summary:A software architecture for hierarchical fuzzy logic hierarchy based on fuzzy logic algorithmic models and separate parallel computing technologies such as MPI and CUDA has been developed, as well as key architectural design solutions to provide the flexibility and performance of a parallel fuzzy system. The patterns of designing and programming of different levels of abstraction, as well as the main software modules and interfaces of the fuzzy software system are analyzed. The method of containerization of applications for the GPU for their reuse and deployment on the selected software platform of high-performance parallel computing is presented.Problems in programming 2018; 2-3: 099-0108
ISSN:1727-4907
DOI:10.15407/pp2018.02.099