Supercomputing: Concepts
Figure 1
Amdahl’s law and limits on scalability (image
from Wikipedia)
Figure 2
Cray-1 Supercomputer at Bell Labs
(Circa-1976)
Figure 3
With a peak performance of 2.79 exaFLOPS, El
Capitan comprises more than 11,000 compute nodes and provides Lawrence
Livermore National Laboratory with a flagship machine 22 times more
powerful than its previous fastest supercomputer, Sierra. (Photos: Garry
McLeod/LLNL)
Figure 4
Figure 5
Compute node (front and back)
Figure 6
Compute node (front)
Figure 7
Compute node (top)
Figure 8
Microprocessor Trends (Performance, Power and
Cores) [1971-2021]
The Life Cycle of a Job
Figure 1
Terminal-based text editor nano
Figure 2
Terminal-based text editor nano
Command Line Interface: The Shell
Figure 1
File Tree
Terminal-based Text Editors: nano, emacs and vim
Figure 1
emacs Reference Card (page 1)
Figure 2
emacs Reference Card (page 2)
Figure 3
Figure 4
Data Transfer: Globus
Figure 1
Open OnDemand - Dashboard
Figure 2
Open OnDemand - File Manager
Figure 3
Globus Online - File Transfer App
Figure 4
Globus Online Web Page
Figure 5
Globus Online Collections
Workload Manager: Slurm
Figure 1
Why do supercomputers use queueing?
Figure 2
NewJob
Figure 3
JobOptions
Software on HPC Clusters: modules, conda and containers
Terminal Multiplexing: tmux
Figure 1
tmux logo
Figure 2
Example of a tmux window
Figure 3
tmux cheat sheet