CACHE REPLACEMENT ALGORITHM BASED ON POPULARITY PREDICTION IN CONTENT DELIVERY NETWORKS

Published 31 Jan 2019 •  vol 122  • 


Authors:

 

Hatim Cherkaoui, Laboratory of Information Processing and Decision Support, University of Sultan Moulay Slimane, Morocco

Driss Ait Omar, Laboratory of Information Processing and Decision Support, University of Sultan Moulay Slimane, Morocco

Mohamed Baslam, Laboratory of Information Processing and Decision Support, University of Sultan Moulay Slimane, Morocco

Mohamed Fakir, Laboratory of Information Processing and Decision Support, University of Sultan Moulay Slimane, Morocco

Abstract:

 

In Order to improve the Quality of Service in Content Delivery Networks for the end users, the choice of an efficient strategy of caching is an asset. Caching is at the heart of content delivery network (CDN) services. Caching works by selectively storing website files on a CDN’s cache proxy servers, a CDN moves websites content to powerful proxy servers optimized for accelerated content distribution. This paper focuses on a predictive strategy of caching based on forecasting number of requests on video content to estimate popularity in future. For that purpose, models called experts, compute future demand value of each video content. Request values generated by each expert are evaluated by a loss function in aim to discuss the accuracy of expert’s prediction. The main goal is to build an efficient caching algorithm that compares to other benchmark algorithms such as LRU and LFU in term of Hit Ratio and Update Ratio. After a performance evaluation, we showed that our proposed approach is better than the existing methods.

Keywords:

 

CDN, Popularity Prediction, Caching Content, Double Exponential Smoothing, LFU, LRU

References:

 

[1] G. Pallis et A. Vakali, “Insight and Perspectives for Content Delivery Networks”, Commun. ACM, vol. 49, no 1, p. 101–106, janv. 2006.
[2] G. Peng, “CDN: Content Distribution Network”, arXiv:cs/0411069, nov. 2004.
[3] V. K. Adhikari, S. Jain, Y. Chen, et Z. Zhang, “Vivisecting YouTube: An active measurement study”, in 2012 Proceedings IEEE INFOCOM, 2012, p. 2521‑2525.
[4] N. B. Hassine, D. Marinca, M. Pascale, et D. Barth, “Machine Learning and Popularity Prediction of a Video Content”, présenté à The 4th International Conference on Performance Evaluation and Modeling in Wired and Wireless Networks (PEMWN), 2015.
[5] D. Ait Omar, M. El Amrani, M. Baslam, et M. Fakir, “A Game-Theoretic Approach for the Internet Content Distribution Chain”, in Networked Systems, 2019, p. 270‑285.
[6] H. Garmani, M. Baslam, et M. Jourhmane, “Caching Games between ISP in Information Centric Network”, International Journal of Control and Automation, vol. 11, no 4, p. 125‑142, avr. 2018.
[7] H. Garmani, M. Outanoute, M. Baslam, et M. Jourhmane, “New Competition-Based Approach for Caching Popular Content in ICN”, in Networked Systems, 2019, p. 286‑300.
[8] K. Arora et D. R. Ch, Web Cache Page Replacement by Using LRU and LFU Algorithms with Hit Ratio: A Case Unification.
[9] G. Pallis, A. Vakali, et E. Sidiropoulos, “FRES-CAR: An Adaptive Cache Replacement Policy”, in International Workshop on Challenges in Web Information Retrieval and Integration, 2005, p. 74‑81.
[10] G. Pallis, K. Stamos, A. Vakali, D. Katsaros, et A. Sidiropoulos, “Replication Based on Objects Load under a Content Distribution Network”, in 22nd International Conference on Data Engineering Workshops (ICDEW’06), 2006, p. 53‑53.
[11] N. B. Hassine, D. Marinca, P. Minet, et D. Barth, “Caching strategies based on popularity prediction in content delivery networks”, in 2016 IEEE 12th International Conference on Wireless and Mobile Computing, Networking and Communications (WiMob), 2016, p. 1‑8.
[12] D. Ait Omar, M. Outanoute, M. Baslam, M. Fakir, et B. Bouikhalene, “On understanding price-QoS war for competitive market and confused consumers”, Computing, juin 2018.
[13] D. Ait Omar, M. Outanoute, M. Baslam, M. Fakir, et B. Bouikhalne, “Joint Price and QoS Competition with Bounded Rational Customers”, in Networked Systems, 2017, p. 457‑471.
[14] M. E. Amrani, H. Garmani, M. Baslam, B. Minaoui, et R. E. Ayachi, “Cournot Game of ISP under Net Neutrality”, International Journal of Control and Automation, vol. 11, no 7, p. 16, 2018.
[15] H. Garmani, M. Baslam, et M. Jourhmane, “Analysis of Competition between CPs Fronting Advertisers”, International Journal of Advanced Science and Technology, vol. 117, p. 53‑66, août 2018.

Citations:

 

APA:
Cherkaoui, H., Ait Omar, D., Baslam, M., & Fakir, M., (2019). Cache Replacement Algorithm Based On Popularity Prediction in Content Delivery Networks. International Journal of Advanced Science and Technology (IJAST), ISSN: 2005-4238(Print); 2207-6360 (Online), NADIA, 122, 35-48. doi: 10.33832/ijast.2019.129.04.

MLA:
Cherkaoui, Hatim, et al. “Cache Replacement Algorithm Based On Popularity Prediction in Content Delivery Networks.” International Journal of Advanced Science and Technology, ISSN: 2005-4238(Print); 2207-6360 (Online), NADIA, vol. 122, 2019, pp. 35-48. IJAST, http://article.nadiapub.com/IJAST/Vol122/4.html.

IEEE:
[1] H. Cherkaoui, D. Ait Omar, M. Baslam and M. Fakir, “Cache Replacement Algorithm Based On Popularity Prediction in Content Delivery Networks.” International Journal of Advanced Science and Technology (IJAST), ISSN: 2005-4238, NADIA, vol. 122, pp. 35-48, Jan. 2019.