Minimizing Task Completion Time in the Cloud based on Random Neural Network

With the development of IoT and 5G, the number of devices accessing the Internet is increasing every day. While mobile edge computing effectively reduces the pressure on cloud centers, cloud centers still face the challenge of task scheduling and resource allocation for a large amount of SaaS applic...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2021 International Conference on Computer, Blockchain and Financial Development (CBFD) s. 60 - 65
Hlavní autoři: Yu, Wang, Tongtong, Wu
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 01.04.2021
Témata:
Cloud computing
Heuristic algorithms
Multi-access edge computing
Neural networks
Processor scheduling
queue theory
random neural network
Simulation
Stochastic processes
task completion time
task dispatching
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:With the development of IoT and 5G, the number of devices accessing the Internet is increasing every day. While mobile edge computing effectively reduces the pressure on cloud centers, cloud centers still face the challenge of task scheduling and resource allocation for a large amount of SaaS applications. In this paper, the conditions for minimizing the average task completion time are derived by a simplified queuing model and an adaptive dynamic scheduling algorithm for minimizing the average task completion time is proposed in combination with stochastic neural networks, which is based on online measurements and takes up very little resources and computation. A diverse range of algorithms are tested in many different environments as a way to analyze algorithm performance. The simulation results show that our proposed algorithm is effective in reducing the average task completion time in a variety of environments.
DOI:10.1109/CBFD52659.2021.00019