Oct 6 – 7, 2022
Zoom
Europe/Berlin timezone

For further inquiries, please contact:

Scientific Program

Introduction

Basic processor and core architecture

  • Out-of-order execution
  • Instruction scheduling
  • Throughput and latency of instructions
  • Critical path and loop-carried dependencies

Introduction to the x86-64 Instruction Set Architecture (ISA)

  • Understanding scalar and vectorized assembly code

Performance analysis of simple kernels

  • Example: STREAM Triad on Intel Ice Lake
  • Hands-on: Dot product on Intel Ice Lake

Introduction to the Open-Source Architecture Code Analyzer (OSACA)

  • How to use OSACA
  • How to use the Compiler Explorer
  • Analyzing kernels using OSACA to find potential bottlenecks
  • Hands-on: PI by integration on Intel Ice Lake

In-core analysis with an Arm ISA

  • Fujitsu A64FX core architecture
  • AArch64 ISA introduction
  • Understanding scalar and vectorized Arm assembly

Case studies: In-core performance engineering on A64FX

  • Sparse Matrix-Vector (SpMV) Multiplication on A64FX
  • Domain Wall kernel from Lattice Quantum Chromodynamics (QCD) on A64FX

Hands-On: 2D Gauss-Seidel on Intel Ice Lake