Introduction to High Performance Computing for Life Scientists

Key Points

Welcome
LECTURE: Measuring Parallel Performance
  • As the number of cores are increased, performance rarely scales linearly.

  • Scalibility changes as the size and complexity of the system is changed.

  • Scalibility can be meaningless if the work load is not distributed evenly among processors.

PRACTICAL: Benchmarking Molecular Dynamics Performance Using GROMACS 1
  • Larger systems scale better to large core-/node-counts than smaller systems.

LECTURE: Computational Building Blocks: Software
  • Most HPC systems run UNIX or Linux as an OS.

  • Each application is a seperate process; the OS will schedule any process.

  • Threads are sub-processes that compose a process.

LECTURE: Computational Building Blocks: Hardware
  • To be determined

PRACTICAL: Benchmarking Molecular Dynamics Using GROMACS 2
  • Hybrid MPI with OpenMP does affect performance.

  • When running hybrid jobs, placement across NUMA regions is important.

LECTURE: Parallel Programming Models
  • To be determined

LECTURE: Compiling software: from source code to executable
LECTURE: Compiling software: from source code to executable
REVIEW: Review of Day 2