Optimasi Scheduling untuk Proses Extract, Transform, Load (ETL) pada Data Warehouse Menggunakan Metode Round Robin Data Partitioning (Studi Kasus: Universitas XYZ)

Agung Yudha Berliantara (Fakultas Ilmu Komputer, Universitas Brawijaya)
Satrio Agung Wicaksono (Fakultas Ilmu Komputer, Universitas Brawijaya)
Aryo Pinandito (Fakultas Ilmu Komputer, Universitas Brawijaya)

Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Vol 1 No 11 (2017): November 2017

The data in the data warehouse can be extracted from many sources through the ETL (extract, transform, and load) processes. The scheduling process in ETL data warehouse is a process that must be passed for the establishment of data warehouse. Problems that arise when performing the scheduling process for the ETL process are to perform the ETL process for very large data, it will require a very large time cost as well. If it is not handled properly, it will probably take a very long time for execution and inconsistency data. The ETL process in the conventional way (not using the Round-Robin method) will be tested and compared with the ETL process using the Round-Robin method to obtain test results for this research. The difference between these two processes is the number of partition tables used in the ETL process. In the ETL process of the Round-robin method, the table target will be partitioned into sections, whereas for the conventional way there is no need to partitioning the table target. After the trial is complete, the difference between the two methods is found in terms of execution time. Through the experiment, it is concluded that the Round-robin method gives a more efficient execution time up to 60,1% depending on the amount of data and the number of partitions used.