A Database Middleware Architecture for MPI-based Cloud Platform

Author

Lin Zhu ; Yucheng Guo

Author_Institution

Dept. of Comput. Sci., Wuhan Univ. of Technol., Wuhan, China

fYear

2014

fDate

24-27 Nov. 2014

Firstpage

107

Lastpage

110

Abstract

To solve the poor performance of existing cloud platforms in high-performance computing, our research team developed a high-performance cloud computing platform based on MPI. As a cloud platform, massive data storage is a key technology. In the research, we mainly studied how to build a MPI cloud platform that supports massive data storage, and implement a distributed cluster constructed by MySQL. This paper introduces the feature of MPI cloud platform, how to build the cloud platform database and the implementation of related SQL operations on the database. Based on this, we test the related operations in the database, and elaborate the shortcomings and challenges of the distributed database in this cloud computing platform.

Keywords

SQL; cloud computing; message passing; middleware; relational databases; MPI-based cloud platform; MySQL; SQL operations; Structured Query Languages; database middleware architecture; distributed cluster; high-performance cloud computing platform; massive data storage; message passing interface; Computer architecture; Distributed databases; IP networks; Memory; Middleware; Semantics; MPI; MySQL clusters; cloud computing; distributed database;

fLanguage

English

Publisher

ieee

Conference_Titel

Distributed Computing and Applications to Business, Engineering and Science (DCABES), 2014 13th International Symposium on

Conference_Location

Xian Ning

Print_ISBN

978-1-4799-4170-4

Type

conf

DOI

10.1109/DCABES.2014.25

Filename

6999067