The computational science community is on the cusp of a new era in high-performance computing. Our objective was to create a place for students to work with multiprocessing and molecular simulations, which requires high computing resources. In addition, we wanted to use our university resources as efficiently as possible. We used a simple simulation class at university and design a dynamic cluster, as clusters have clear price/performance, user interface, and maintenance advantages over traditional supercomputers [1].
Cluster objectives: a) to make sure students can use parallel processing and molecular simulations on the cluster and it will not affect ordinary usage of the computer class. b) automatic maintenance and monitoring of this dynamic cluster. c) cluster modifications for optimal performance. d) ability to work from home.