Loading [MathJax]/extensions/TeX/color_ieee.js
Predictive Storage Management for Cloud-Based Video Streaming Using ML ARIMA Model | IEEE Conference Publication | IEEE Xplore

Predictive Storage Management for Cloud-Based Video Streaming Using ML ARIMA Model


Abstract:

The exponential increase in video streaming has overburdened data centers, making efficient video streaming in the cloud essential. One potential solution is the tradeoff...Show More

Abstract:

The exponential increase in video streaming has overburdened data centers, making efficient video streaming in the cloud essential. One potential solution is the tradeoff between transcoding and storing a video stream, as storing operations are cheaper than transcoding. This research explores the use of machine learning ARIMA algorithms to predict the number of views for each video in the repository for the next period of time. This enables informed decisions about which videos to keep or delete from the cloud, ultimately reducing cloud service costs. The study evaluates the accuracy of the ARIMA model in predicting future views for three datasets of varying sizes and levels of variability. Our results show 82% of predicting accurate views, indicating that the ARIMA model can effectively capture the overall trend of video views, providing valuable insights for service providers looking to optimize their storage management practices and reduce costs. This research contributes to the development of more cost-efficient video streaming in the cloud, which is essential for meeting the increasing volume of video stream traffic. By leveraging the ARIMA model, cloud-based platforms can make informed decisions on storage management, enhancing their overall service quality and user experience.
Date of Conference: 06-09 August 2023
Date Added to IEEE Xplore: 31 January 2024
ISBN Information:

ISSN Information:

Conference Location: Tempe, AZ, USA

References

References is not available for this document.