ANALISIS KOMPUTASI PARALEL DAN SERIAL PADA ALGORITMA MERGE SORT

Abstract

This paper describes the application of parallel computing technologies in the merge sort algorithm. The data used in this study as many as 100 to 6 million. Data in the form of randomly generated numbers. In the testing phase, parallel computing processor = 2, processor = 4 (computer simulation), and 2 computers connected in a peer to peer. The results showed that the amount of data that is not too big, serial computation time is faster when compared to the parallel computing time. By using two computer peer to peer the amount of data that can reach 6 million sorted data. The average of Speed ​​Up value of  p = 2, p = 4, and computer = 2 was 2.273738, 1.980198 and 1.074942.

Keywords : merge sort, peer to peer, speed up