What have we learnt from using real parallel machines to solve real problems?
C3P Proceedings of the third conference on Hypercube concurrent computers and applications - Volume 2
High Performance Cluster Computing: Programming and Applications
High Performance Cluster Computing: Programming and Applications
| Hi-index | 0.01 |
In modern organization information and data are very importance aspects to be concerned on every application. Most of applications store data and information using database management system. Those data and information typically are stored in a centralized storage that makes easy for administrator to manage and manipulate those data. However this method is limited by the capacity of database server and the way of processing approach. The database server will handle all clients' requests that consume performance of CPU power, memory bandwidth, I/O bandwidth, and network bandwidth. Therefore database server should have high capacities in term of CPU power, memory, network bandwidth and large storage that will be costly in budget. Hence there are many researches that propose some other approaches to improve system performance. For example, DBMS such as SQL server and Oracle utilize distributed processing of database using several database servers but the data itself still stored in a centralized storage as shared storage. Currently, MySQL has launched MySQL Cluster DBMS that supports the distribution of database and distributed processing. The data of this methodology will be stored in each storage nodes in the cluster by using hashing function. In this research we would like to evaluate the performance of distributed data using MySQL Cluster DBMS together with our PC Cluster computers. We hope the distribution of data and using cluster technology would improve the performance of database processing.