TY - GEN
T1 - A Machine Learning based Context-aware Prediction Framework for Edge Computing Environments
AU - Aljulayfi, Abdullah Fawaz
AU - Djemame, Karim
N1 - Publisher Copyright:
Copyright © 2021 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved
PY - 2021
Y1 - 2021
N2 - A Context-aware Prediction Framework (CAPF) can be provided through a Self-adaptive System (SAS) resource manager to support the autoscaling decision in Edge Computing (EC) environments. However, EC dynamicity and workload fluctuation represent the main challenges to design a robust prediction framework. Machine Learning (ML) algorithms show a promising accuracy in workload forecasting problems which may vary according to the workload pattern. Therefore, the accuracy of such algorithms needs to be evaluated and compared in order to select the most suitable algorithm for EC workload prediction. In this paper, a thorough comparison is conducted focusing on the most popular ML algorithms which are Linear Regression (LR), Support Vector Regression (SVR), and Neural Networks (NN) using real EC dataset. The experimental results show that a robust prediction framework can be supported by more than one algorithm considering the EC contextual behavior. The results also reveal that the NN outperforms LR and SVR in most cases.
AB - A Context-aware Prediction Framework (CAPF) can be provided through a Self-adaptive System (SAS) resource manager to support the autoscaling decision in Edge Computing (EC) environments. However, EC dynamicity and workload fluctuation represent the main challenges to design a robust prediction framework. Machine Learning (ML) algorithms show a promising accuracy in workload forecasting problems which may vary according to the workload pattern. Therefore, the accuracy of such algorithms needs to be evaluated and compared in order to select the most suitable algorithm for EC workload prediction. In this paper, a thorough comparison is conducted focusing on the most popular ML algorithms which are Linear Regression (LR), Support Vector Regression (SVR), and Neural Networks (NN) using real EC dataset. The experimental results show that a robust prediction framework can be supported by more than one algorithm considering the EC contextual behavior. The results also reveal that the NN outperforms LR and SVR in most cases.
KW - Edge Computing
KW - Linear Regression
KW - Machine Learning
KW - Neural Networks
KW - Prediction Framework
KW - Self-adaptive Systems
KW - Sliding Window
KW - Support Vector Regression
UR - http://www.scopus.com/inward/record.url?scp=85137949683&partnerID=8YFLogxK
U2 - 10.5220/0010379001430150
DO - 10.5220/0010379001430150
M3 - Conference contribution
AN - SCOPUS:85137949683
T3 - International Conference on Cloud Computing and Services Science, CLOSER - Proceedings
SP - 143
EP - 150
BT - CLOSER 2021 - Proceedings of the 11th International Conference on Cloud Computing and Services Science
A2 - Helfert, Markus
A2 - Ferguson, Donald
A2 - Pahl, Claus
PB - Science and Technology Publications, Lda
T2 - 11th International Conference on Cloud Computing and Services Science, CLOSER 2021
Y2 - 28 April 2021 through 30 April 2021
ER -