Home // CLOUD COMPUTING 2015, The Sixth International Conference on Cloud Computing, GRIDs, and Virtualization // View article

Cloud Storage Prediction with Neural Networks

Authors:
Stefan Frey
Simon Disch
Christoph Reich
Martin Knahl
Nathan Clarke

Keywords: Cloud Computing, Storage, Prediction, Neural Net- works, SLA, QoS

Abstract:
In this work, we present an Artificial Neural Network approach to predict the usage, size and type of a cloud storage to enable better compliance with Service Level Agreements (SLAs). One of the biggest advantage of cloud infrastructures is scalability on demand. Cloud services are monitored and based on utilization and performance need, they get scaled up or down, by provision or deprovision of resources. The goal of the presented approach is to predict and thereof select the right amount of storage with a minimum of preallocated resources, as well as the corresponding storage type based on the predicted performance needs in order to reduce SLA violations. Evaluation of the results obtained by simulation confirm that, by using this approach, SLA violations decreased compared to a threshold value control system.

Pages: 52 to 56

Copyright: Copyright (c) IARIA, 2015

Publication date: March 22, 2015

Published in: conference

ISSN: 2308-4294

ISBN: 978-1-61208-388-9

Location: Nice, France

Dates: from March 22, 2015 to March 27, 2015