Doctoral Seminar: Touraj Laleh

Speaker: Touraj Laleh

Supervisors: Drs. J. Paquet, Y. Yan

Supervisory Committee:
Drs. R. Glitho, V. Haarslev, R. Witte

Title:  A Context-Aware Service Brokerage for SLA-based Service Provisioning in Cloud Environment

Date: Thursday, March 31, 2016

Time: 11:40am

Place: EV 3.309

ABSTRACT

Context-aware services use contextual information to provide more customized services for their clients. However, there is always a gap between clients’ needs in getting more customized services and the capability of abstract SaaS providers to respond according to the clients’ and providers’ contexts. Cloud services brokerage (CSB) is a model where a businesses or other entities add value to one or more cloud services on behalf of one or more consumers of those services via three primary brokerage roles: aggregation, integration and customization. However, there are still many issues in terms of integration and service level agreement (SLA) management not only among services used by a specific application (client), but also among services and data of multiple applications.

In this research we propose a service brokerage as an intermediation approach among clients and providers designed to manage various aspects of service provisioning including service selection, service composition, and SLA management. We also propose a mechanism to apply effects of data integration in terms of policies in service composition. Finally, we study the effects of our approach in cloud environment. For the evaluation phase we have developed initial test bed setup to validate our algorithm and the approach (in progress) in general and conduct various experiments on our defined scenarios. To scale out our tests and experiments, we are in the process of integration of GIPSY's general education engine (GEE), a distributed multi-tier architecture for evaluation of programs in a distributed and demand-driven manner to simulate cloud
environment with large amount of demands and services where it is easier to setup a large collection of nodes and service demands.