Clusters and Massively Parallel Machines

This chapter highlights clusters and massively parallel machines architecture. Both of these architectures are referred as being loosely-coupled (processors in one node do not have any close link with the processors in another node). Processors in different nodes do not share memory and each node runs its own copy of the operating system. A cluster includes reasonable number of systems, which are interconnected to share resources commonly known as cluster resources. Each system in a cluster is called a node. Each node is a real, complete system, with all the resources it needs, such as processor, memory, I/O devices, and storage subsystem generally working under the control of its own copy of the operating system. Massively parallel machines are much advantageous than traditional supercomputers in terms of cost as they are designed and manufactured using standard component available, while supercomputers tends to use niche technologies, such as GaAs (Gallium Arsenide).