PROJECT TITLE :
Managing Performance Interference in Cloud-Based Web Services
Net services have increasingly begun to rely on public cloud platforms. The virtualization technologies utilized by public clouds can, however, trigger contention between virtual machines (VMs) for shared physical machine resources, thereby resulting in performance problems for Internet services. Past studies have exploited physical-machine-level performance metrics like clock cycles per instruction to detect such platform-induced performance interference. Unfortunately, public cloud customers do not have access to such metrics. They'll only sometimes access VM-level metrics and application-level metrics such as transaction response times, and such metrics alone are often not useful for detecting inter-VM rivalry. This poses a tough challenge to Internet service operators for detecting and mitigating platform-induced performance interference problems within the cloud. We propose a machine-learning-primarily based interference detection technique to deal with this drawback. The technique applies collaborative filtering to predict whether or not a given transaction being processed by a Web service is adversely plagued by interference. The results will be then used by a management controller to trigger remedial actions, e.g., reporting problems to the system manager or switching cloud suppliers. Results employing a realistic Web benchmark show that the approach is effective. The foremost effective variant of our approach is in a position to detect about 96p.c of performance interference events with almost no false alarms. Furthermore, we show that a load redistribution technique that exploits the data from our detection technique is able to more effectively mitigate the interference than techniques that are interference agnostic.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here