PhD. Seminar #2 by:

Peng Cheng

 

Date: Wednesday Auguest 21, 2019

Time: 11:00am-12:30pm

Location: 3105, Lambton Tower

 

To ensure the performance of software systems based upon service-oriented architecture (SOA), quality of service (QoS) is documented as a service-level agreement (SLA) with official commitments specified for both service consumers and providers to follow. Since monetary penalties for violation are often included in the SLA,  service providers have a strong interest in keeping their commitments to avoid and reduce the situations that may cause SLA violations. In recent years, many research works are devoted to predict, avoid and reduce the SLA violation. However, once violation of SLA occurred, these existing methods are not capable of optimizing QoS or have the issue of over-provisioning resource. As part of an overall framework developed to predict and prevent SLA violation, this paper proposes a novel strategy that uses combined global and local searching to help service providers in achieving design optimum without over-provision of resources. Experimental results demonstrate that the proposed method outperforms other commonly used methods in search efficiency while retaining competitive with those methods in optimization accuracy.

Thesis Committee:

Seminar Announcement