PPO - An Approach to Personalized Web Acceleration

Shailesh Shivakumar; P.V Suresh

Shailesh Shivakumar Indira Gandhi National Open University India
P.V Suresh Indira Gandhi National Open University India

Abstract

A responsive web page plays critical role in the overall success of the online channels. It directly impacts user experience and also influences the search engine rankings. Most of the web systems use personalized web to provide relevant and contextual information to web users. Personalized web provide content, data and functionality based on various personalization parameters such as user interests, user preferences, user profile attributes, location, device, user navigation patterns, purchase behavior and such. Personalized web is a key business enabler to drive the user traffic and keep the web users engaged by providing useful information. One of the main side effects of personalized web is on the performance. Traditional performance optimization techniques cannot be scaled and reused for personalized web due to the dynamics of the personalized content and due to security/privacy concerns. In addition to relevant content, web users would also expect good performance in personalization scenarios. Web architecture needs to design for optimal performance in personalization scenarios for long-term success of web systems. In this paper we have tried to address this crucial issue by discussing various aspects of personalized performance optimization algorithms. We have discussed a novel approach using â€œPersonalization performance Optimizationâ€ (PPO) framework that has resulted in 30% increase in page response times and 35% increase in cache hit ratio during our experiments.

References

[1] Web performance optimization: http://en.wikipedia.org/wiki/Web_performance_optimization [2] For Impatient Web Users, an Eye Blink Is Just Too Long to Wait: http://www.nytimes.com/2012/03/01/technology/impatientweb-users-flee-slow-loading-sites.html?_r=2 [3] Akamai Report: http://www.akamai.com/html/about/press/releases/2009/press_091409.html [4] Speed Is A Killer â€“ Why Decreasing Page Load Time Can Drastically Increase Conversions: http://blog.kissmetrics.com/ speed-is-a-killer/. [5] Souders. S â€“ Even Faster Web Sites: Performance Best Practices for Web Developers; Oâ€™Reilly Media, 2009 [6] Souders. S â€“ High Performance Web Sites: Essential Knowledge for Front-End Engineers; Oâ€™Reilly Media, 2007 [7] Best Practices for Speeding Up Your Web Site: http://developer.yahoo.com/performance/rules.html [8] Web Performance Best Practices: http://code.google.com/speed/page-speed/docs/rules_intro.html [9] WPO â€“ Web Performance Optimization: http://www.stevesouders.com/blog/2010/05/07/wpo-web-performance-optimization/ [10] Stefanov. S â€“ Web Performance Daybook; Oâ€™Reilly Media, 2012 [11] Grioen, James., Appleton, Randy. (1994). Reducing File System Latency using a Predictive Approach, In: Proceedings of the 1994 Summer USENIX Technical Conference, Cambridge MA, (June). [12] Padmanabhan, V. N., Mogul, J. C. (1996). Using predictive prefetching to improve World Wide Web latency. ACM SIGCOMM Computer Communication Review. [13] Gu, Peng., Wang, Jun., Zhu, Yifeng., Jiang, Hong., Shang, Pengju. (2010). A Novel Weighted-Graph-Based Grouping Algorithm for Metadata Prefetching. CSE Journal Articles. Paper 44. [14] Kotz, D., Ellis, C. S. (1993). Practical Prefetching Techniques for Multiprocessor File Systems, J. Distributed and Parallel Databases . 1 (1) 33-51, Jan. 1993 [15] Lei, H., Duchamp, D. (1997). An Analytical Approach to File Prefetching, In: Proc. USENIX Ann. Technical Conf., (January). [16] Lee, H., An, B., Kim, E. (n.d.). Adaptive Prefetching Scheme Using Web Log Mining in Cluster-Based Web Systems. 2009 IEEE International Conference on Web Services. [17] Dahlan, A., Nishimura, T. (n.d.). Implementation of asynchronous predictive fetch to improve the performance of Ajaxenabled web applications, In: Proceedings of the 10th International Conference on Information Integration and Web-based Applications & Services - IiWAS â€˜08. [18] Yang, Q., Zhang, H., Li, T. (n.d.). Mining web logs for prediction models in WWW caching and prefetching, In: Proceedings of the Seventh ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD â€˜01. [19] Venkataramani, A., Yalagandula, P., Kokku, R., Sharif, S., Dahlin, M. (n.d.). The potential costs and benefits of long-term prefetching for content distribution. Computer Communications, 367-375. [20] Bouras, C., Konidaris, A., Kostoulas, D. (n.d.). Predictive Prefetching on the Web and Its Potential Impact in the Wide Area. World Wide Web, 143-179. [21] Xu, C., Ibrahim, T. (n.d.). Towards semantics-based prefetching to reduce Web access latency. 2003 Symposium on Applications and the Internet, 2003. Proceedings. [22] Frelechoux, L., Kamba, T. (1997). An architecture to support personalized Web applications. Computer Networks and ISDN Systems, 29. Chicago [23] Reda, Azarias., Cutrell, Edward., Noble, Brian. 2011. Towards improved web acceleration: leveraging the personal web. In: Proceedings of the 5th ACM workshop on Networked systems for developing regions (NSDR â€™11). ACM, New York, NY, USA, 57-62. [24] Douglis, F., Haro, A., Rabinovich, M. (1997), December. HPP: HTML Macro-Preprocessing to Support Dynamic Document Caching. In: USENIX Symposium on Internet Technologies and Systems p. 83-94. [25] Ravi, J. (2009). A survey on dynamic Web content generation and delivery techniques. J Network Comput Appl. [26]de la Ossa, B., Gil, J. A., Sahuquillo, J., Pont, A. (2007). Web prefetch performance evaluation in a real environment.In: Proceedings of the 4th international IFIP/ACM Latin American conference on Networking (LANC â€™07). ACM, New York, NY, USA, 65-73. [27] Palpanas, T., Mendelzon, A. (1999). Web prefetching using partial match prediction, In: Proc. of the 4th International Web Caching Workshop, San Diego, USA. [28] Fan, L., Cao, P., Lin, W., Jacobson, Q. (1999). Web prefetching between low-bandwidth clients and proxies: potential and performance. In: ACM SIGMETRICS.