Supercomputers for Starters
Initializing search
Supercomputers for Starters
Table of contents
Preliminary words
Introduction
Introduction
Goals
Why supercomputing?
What it is not
A compartmentalised supercomputer
Overview of the notes
Processors for supercomputers
Processors for supercomputers
Introduction
A basic CPU
ILP I - Pipelining
ILP II - Superscalar
Vector computing
Shared memory
Distributed memory
A modern (super)computer
Node architecture
Fast (past) evolution
An important difference...
Processors: Lessons learnt
Further reading on processors
The memory hierarchy
The memory hierarchy
The memory performance gap
The memory pyramid
AMD Rome
AMD Milan
Storing data on Supercomputers
Storing data on Supercomputers
Introduction
Problems
Parallel file systems
A storage revolution?
To remember
Further reading on storage
Putting it all together
Putting it all together
Scaling of technology
Dennard scaling
Transistor cost
3 keywords
Andy and Bill's law
Software, not hardware
Building up the supercomputer
Middleware
Middleware
Middleware: Introduction
Shared memory programming
Vector programming
Distributed memory programming
Middleware: Examples
Further reading on middleware
What can we expect?
What can we expect?
Speed-up
Matrix multiplication
Speed-up and efficiency
Conclusions
Accelerators
Accelerators
What are accelerators?
Offloading
CPUs with accelerator features
Accelerator programming
Status of GPU computing
Further reading on accelerators
Conclusions
Conclusions
Tags index
404 - Not found
Back to top