ABSTRACT
Cloud computing implies that computing is not only operated on local computers, but on centralized facilities by third-party computing and storage utilities. It refers to both the applications delivered as services over the Internet and system hardware/- software in datacenters as service providers. Cloud solutions seem to state master keys for the IT enterprises which suffer from budget concerns and economic woes, and a number of industry projects have been started to create a global, multi-data center, open source cloud computing testbed for industry, research, and education. Encouraging opportunities also brings out corresponding challenges. Cloud com-
puting is easily confused with several existing technologies including grid computing, utility computing, web service, and virtualization. Again, cloud computing is a newly evolved delivery model. It covers with equal importance both technology and business requirements, and it lets users focus on their abilities on demand by abstracting its technology layer. In that case, the scheduling problem in cloud computing is worth reconsidering by researchers and engineers. In this book, we addressed the resource allocation problem in terms of economic aspects to meet business requirements. At the same time, we were concerned with the real-time schedulability test to provide the cloud datacenter with technical supports. Both theoretical and practical efforts were made to solve cloud scheduling problems and to facilitate the succeeding researches. In order to utilize cloud computing to serve as the infrastructure of multi-
dimensional data analysis applications, the combination of traditional parallel database optimization mechanisms and cloud computing is expected. In this book, we try to utilize methods of cloud computing to satisfy commercial software requirements. In addition to realizing a concrete multi-dimensional data analysis query with MapReduce, we mostly focus on the performance optimization, and combine MapReduce with several optimization techniques coming from parallel database. It is an important aspect in designing cloud computing-based solution for business software. This approach does not depend on a third-party product, and it can guarantee the performance application. Besides the mentioned contributions, our work has raised many interesting ques-
tions and issues that deserve further research.
