Browsing by Author "Cunha, Carlos Augusto da Silva"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
- Reboot-based Recovery of Performance Anomalies in Adaptive Bitrate Video-Streaming ServicesPublication . Cunha, Carlos Augusto da Silva; Silva, Luis M. ePerformance anomalies represent one common type of failures in Internet servers. Overcoming these failures without introducing server downtimes is of the utmost importance in video-streaming services. These services have large user abandon- ment costs when failures occur after users watch a significant part of a video. Reboot is the most popular and effective technique for overcoming performance anomalies but it takes several minutes from start until the server is warmed-up again to run at its full capacity. During that period, the server is unavailable or provides limited capacity to process end-users’ requests. This paper presents a recovery technique for performance anomalies in HTTP Streaming services, which relies on Container-based Virtualization to implement an efficient multi-phase server reboot technique that minimizes the service downtime. The recovery process includes analysis of variance of request-response times to delimit the server warm-up period, after which the server is running at its full capacity. Experimental results show that the Virtual Container recovery process completes in 72 seconds, which contrasts with the 434 seconds required for full operating system recovery. Both recovery types generate service downtimes imperceptible to end-users.
- SHStream: Self-Healing Framework for HTTP Video-StreamingPublication . Cunha, Carlos Augusto da Silva; Silva, Luis M.HTTP video-streaming is leading delivery of video content over the Internet. This phenomenon is explained by the ubiquity of web browsers, the permeability of HTTP traffic and the recent video technologies around HTML5. However, the inclusion of multimedia requests imposes new requirements on web servers due to responses with lifespans that can reach dozens of minutes and timing requirements for data fragments transmitted during the response period. Consequently, web- servers require real-time performance control to avoid playback outages caused by overloading and performance anomalies. We present SHStream , a self-healing framework for web servers delivering video-streaming content that provides (1) load admit- tance to avoid server overloading; (2) prediction of performance anomalies using online data stream learning algorithms; (3) continuous evaluation and selection of the best algorithm for prediction; and (4) proactive recovery by migrating the server to other hosts using container-based virtualization techniques. Evaluation of our framework using several variants of Hoeffding trees and ensemble algorithms showed that with a small number of learning instances, it is possible to achieve approximately 98% of recall and 99% of precision for failure predictions. Additionally, proactive failover can be performed in less than 1 second