ANALISIS PERBANDINGAN KINERJA SISTEM CLUSTER PVM DENGAN MPI TERHADAP APLIKASI POV-RAY
| Main Authors: | , Aditia Prana Kusuma, , Teguh Bharata Adji, S.T., M.T., M.Eng., Ph.D, |
|---|---|
| Format: | Thesis NonPeerReviewed |
| Terbitan: |
[Yogyakarta] : Universitas Gadjah Mada
, 2012
|
| Subjects: | |
| Online Access: |
https://repository.ugm.ac.id/99327/ http://etd.ugm.ac.id/index.php?mod=penelitian_detail&sub=PenelitianDetail&act=view&typ=html&buku_id=55679 |
Daftar Isi:
- Animation recently grows very rapidly, shown by the number of animation movies produced. The raise of freeware-licensed rendering applications supports even faster growth, one of them is POV-Ray. However, in order to handle rendering process for some complex 2D or 3D images, considerably long period of time is required. The solution is by using distributed rendering system in clustering system. Choosing clustering systems is also important to consider because it influence the tasks distribution and time required to completion the tasks. This research analyzed the performance of clustering system, in order to gain better clustering system in conducting distributed rendering process. The research was conducted by experiment using ten virtual computers to implement PVM and MPI clustering systems using POV-Ray application. The result of this research shows that on every addition of node on the system the rendering time, speedup, and efficiency rate produced increase faster than only using single node in undertaking rendering task. The average speedup and efficiency rate on PVM clustering system is higher than those of MPI clustering system, as shown at Testing II up top Testing V, in which the speedup on PVM clustering system is 2,94 and its efficiency rate is 84,78% while the speedup of MPI clustering system is 2,44 and its efficiency rate is 68,29%. It can therefore be concluded that PVM clustering system is better than MPI clustering system in term of distributed rendering process.