Getting Started


visiting_hero_web.jpg


Steps to getting started using the Physics Farm system at Jefferson Lab are listed below:

Scientific Computing Resources

Summary of resources at JLab:

  • High Performance Computing (HPC) for LQCD, ~20000 cores, ~180 GPUs
  • Batch Computing for Experimental Physics (the "farm"), ~25000 CPUs
  • Multiple Disk Systems (online storage), ~7 Petabytes
  • The Tape Library for offline storage, ~30 Petabytes
  • Interactive nodes, wide area gateway nodes and infrastructure support nodes

Details are found in the following subsections.

GPU Clusters

logocolor4.png


GPUs provide enormous floating point capacity and memory bandwidth, and can yield up to 10 times as much performance (science) per dollar, but do require the use of specialized libraries or specialized programming techniques.

Intel MPI

logocolor4.png


In this chapter we note the main points about running Intel MPI on the Knights Landing cluster.

KNL System Overview

logocolor4.png

 

Hardware

logocolor4.png


 

LQCD User's Guide

logocolor4.png

 

Farm User's Guide v2


visiting_hero_web.jpg


The Jefferson Lab scientific computing environment for experimental physics is comprised of:

Pages