Research

Research Highlights

"Cloud Broker and Cloudlet for Workflow Scheduling" Published by Professor Chan-Hyun Youn

Professor Chan-Hyun Youn of our department has published a book on deep learning titled “Cloud Broker and Cloudlet for Workflow Scheduling”.

It has been published by Springer Publishing Co. with Professor Min Chen of China and Professor Patrizio Dazzi of Italy. More related information about the book is on below.

 

Title: Cloud Broker and Cloudlet for Workflow SchedulingSpringer, Jul., 2017

Authors: Chan-Hyun Youn, Min Chen, Patrizio Dazzi

Linkhttp://www.springer.com/us/book/9789811050701

 

Summary:

 In this book, we provide a comprehensive overview of the background and related technologies, challenges and applications of cloud broker system, which is the middleware between the user and the resource provider by providing resource allocation service for satisfying users’ requirements. . Especially, we present detailed implementation of cloud broker system to address the great challenges for providing an optimal resource reservation capacity control management: 1) different objective of user-centric Service Level Agreement (SLA), 2) complexity of resource allocation for workflow jobs, 3) uncertainty of the resource demands, 4) difficulty on data scheduling in geographically distributed cloud applications, 5) different resource types and the leasing costs, and 6) different payment plans and leasing duration. Specifically, cloud broker system plays a critical role as an intermediate negotiator between the user and the service provider, which orchestrates the application policy on the user’s side and the resource policy on the service provider’s side. The broker functions to bridge the user with the service provider while hiding many workflow mapping details so that the user can utilize the vast resource pool within the internet in a transparent and easy way. Moreover, the broker can offer different service levels in according with different users’ requirements and provide corresponding Quality of Service (QoS) guarantees respectively. The broker has both the functionalities of workflow management and the resource management.