Failure Prediction in Video-Streaming Servers Through Performance Analysis of Server and Client-Server Interactions



Video-streaming services are nowadays one of the biggest contributors to Internet traffic. Due to the real-time characteristic of video-streaming, very low failure detection and recovery delays are required. Proactive recovery in face of performance degradation is a prominent area to explore. Current performance analysis work in video-streaming focuses mostly on capacity planning and session admission through complex workload and resource modeling. These models have limited application when degradation is not explained by client workload (e.g., dynamic resource reallocation, software faults and misconfiguration). We explore server-side monitoring of performance degradations in video-streaming servers, based on statistical analysis of server metrics and client-server interaction messages. Statistical analysis of event logs show that analyzed metrics can be used as symptoms of failures to anticipate them and thus enabling proactive recovery. Exception is server overloading caused by streaming of unpopular videos, which are exposed by metrics when QoS degradation is close to accepted quality thresholds.


Failure Prediction, Video-Streaming


Failure Prediction in Streaming Servers


Parallel and Distributed Computing and Networks, February 2011

PDF File

Cited by

No citations found