Design of a parallel genetic algorithm for the Internet
Why is this work in the frame?
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
The three-model screen
all 1,000 screened works →All three models called this out of scope.
Design of a parallel genetic algorithm; a computational tool for optimization, not a study of research practice.
It develops a computational algorithm for optimization rather than studying research methods or practice.
Computer science algorithm design for parallel genetic algorithms, not metaresearch on how research is done.
Abstract
This paper proposes a parallel implementation of the genetic algorithm (GA) on the Internet which will improve the algorithm's performance. It is motivated by the possibility of aiding research into complex search and optimization problems that use the GA. Requirements and constraints regarding parallelization of the GA are identified. A parallel GA is developed for an ideal PRAM architecture and is shown to have an asymptotic running time of O(log n), an improvement over the sequential GA. A parallel GA is also designed for a Unix network and has an asymptotic running time comparable to the ideal system. The algorithm is a decentralized, asynchronous, and fault-tolerant design that matches the characteristics of the network. The GA population is divided into colonies that are distributed among processors. Trade policies are executed for the exchange of genes.
Stored with the screening record, where it is evidence for the labels above.
The record
- Venue
- Topic
- Evolutionary Algorithms and Applications
- Field
- Computer Science
- Canadian institutions
- University of Manitoba
- Funders
- —
- Keywords
- Computer scienceUnixGenetic algorithmAsynchronous communicationParallel computingParallel algorithmIdeal (ethics)PopulationThe InternetFault toleranceDistributed computingAlgorithmComputer networkOperating system
- Has abstract in OpenAlex
- yes