PRACE Course: Node-Level Performance Engineering

This course teaches performance engineering approaches on the compute node level. "Performance engineering" as we define it is more than employing tools to identify hotspots and bottlenecks. It is about developing a thorough understanding of the interactions between software and hardware. This process must start at the core, socket, and node level, where the code gets executed that does the actual computational work. Once the architectural requirements of a code are understood and correlated with performance measurements, the potential benefit of optimizations can often be predicted. We introduce a "holistic" node-level performance engineering strategy and apply it to different algorithms from computational science. Architectural details that are relevant for performance, such as pipelining, SIMD, superscalarity, memory hierarchies, etc., are covered in due detail.

The course is a PRACE training event.

• Introduction
  • Our approach to performance engineering
  • Basic architecture of multicore systems: threads, cores, caches, sockets, memory
  • The important role of system topology
• Tools: topology & affinity in multicore environments
  • Overview
  • likwid-topology and likwid-pin
• Microbenchmarking for architectural exploration
  • Properties of data paths in the memory hierarchy
  • Bottlenecks
  • OpenMP barrier overhead
• Roofline model: basics
  • Model assumptions and construction
  • Simple examples
  • Limitations of the Roofline model
• Pattern-based performance engineering
• Optimal use of parallel resources
  • Single Instruction Multiple Data (SIMD)
  • Cache-coherent Non-Uniform Memory Architecture (ccNUMA)
  • Simultaneous Multi-Threading (SMT)
• Tools: hardware performance counters
  • Why hardware performance counters?
  • likwid-perfctr
  • Validating performance models
• Roofline case studies
  • Dense matrix-vector multiplication
  • Sparse matrix-vector multiplication
  • Jacobi (stencil) smoother
• Optional: The ECM performance model