REPLIKASI UNTUK MENINGKATKAN KINERJA DAN KETERSEDIAAN DATA (STUDY KASUS SISTEM INFORMASI AKADEMIK)

Abstract

This study aims to find the algorithms for designing a distributed database architecture with replication method that will be implemented on an academic database. From this study it is hoped that will get a distributed database architecture that can cope with the availability of data on academic information system. By using a distributed database an application can provide information to its end users to retrieve data from the sources closest to the network from the end user transactions.This design was developed and built using the MySQL database type InnoDB. The design was focused in managing a distributed database using the replication method. Transaction processing was only done to the master, while the slave was only a mirror. Data that has been transacted by the slave will be synchronized with the two methods of automatic and manual by following the pattern of events in the academic system. Master was put on the intranet so that the area will ensure the safety of its data, while the slave was spread to multiple database servers, both intranet and public. Web applications that use the data base was placed disuatu domains on the Internet. The results of this study indicate that by doing the exact distribution of the database. With this method an information system is not necessarily connected to a database, but sufficiently connected to the nearest slave database of network users, so in addition to reducing the server load will also further increase the speed in obtaining data and will impact on the confidence level stack holder.