High Performance Computing on Cluster and Multicore Architecture

Abstract

High Performance Computing have several issues on architecture, resources, computational model and data. The challenge is establishing the mature architecture with scalable resources. The cluster architecture and multicore architecture implement to produce high performance on computation and process. This research works on architecture development and performance analysis. The cluster architecture build on Raspberry Pi, a single board computer, implement MPICH2. Raspberry Pi cluster build on Raspbian Wheezy operating system and test by metrics computation applications. The multicore architecture build on single computer with Core i5 and Core i7 architecture. The research use himeno98 and himeno16Large tools to analysis the processor and memory allocation. The test run on 1000x1000 matrices and benchmarked with OpenMP. The analysis focuses on CPU Time, FLOPS, and score. The result show on cluster architecture have 2576,07 sec in CPU Time, 86,96 MLPOS, and 2,69 score. The result on Core i5 architecture have 55,57 sec in CPU time, 76,30 MLOPS, and 0,92 score. The result in Core i7 architecture have 59,56 sec CPU Time, 1427,61 MLOPS, and 17,23 score. The cluster and multicore architecture results show that computing process are effected by architecture models. High performance computing architecture that has been built on this result can give learn on the development of HPC architecture models, and baseline performance. In the future it will use for determine the delivery architecture model on HPC and can be test by more variation of load.