Introduction
What exactly is High Performance Computing (HPC)? Textbook answers say
things like HPC "refers to the full range of supercomputing activities
including existing supercomputer systems, special purpose and
experimental systems, and the new generation of large scale parallel
architectures." If such a statement clears things up for you, then
you probably already know a great deal about HPC. However, if you still
don't have a clue, don't worry, I had the same feeling when I read this
definition initially. This is the greatest problem with virtually
all
HPC resources: their explanations assume previous knowledge that many of
us simply do not have. Thus, lets start at the beginning with the
simplest of explanations.
HPC can be boiled down to one thing: speed. The goal is to achieve the
maximum amount of computations in minimum amount of time. However, from
this goal grows the myriad complexities that make up HPC. The intent of
the author of these pages is to muck through these complexities without
confusing the reader (or himself).
Much of these pages will discuss computer architecture, because computer
architecture is what high performance is derived from. Central to this
topic will be discussions about CISC, RISC, pipelining,
parallelism, and memory.
Finally, we will look at the reason for the complexity that is HPC, its
uses, such as creating simulations to answer real
world questions and quickly completing massive computations.
Next
Home