Home

Building GGP

The following pages give information on GGP's build system and dependencies

• GGP Quick Start Guide

  1. Installation
  2. Internal Testing
  3. Kernel Autotuning
  4. Debugging
  5. Manually Running Basic Tests

• GGP Build With CMake

  1. GGP and CMake
  2. Obtaining CMake
  3. Building GGP using CMake
  4. Reducing GGP build time
  5. Building GGP with clang-cuda
  6. Building GGP with HIP/ROCm

• GGP Environment Variables

Multi-GPU GGP

These pages give information on how to build GGP using MPI, including some machine-specific instructions

• Multi GPU Support
  1. Compiling for Multi-GPU
  2. Running
  3. Running GGP's Tests
  4. Multi-GPU Emulation
  5. Peer-to-peer Communication
  6. GPU Direct RDMA and CUDA-aware MPI
  7. Maximizing GDR Performance
  8. Dependence on CUDA_DEVICE_MAX_CONNECTIONS
  9. Low-level Details
  10. Dslash Policy Tuning
  11. Dslash Component Benchmarking
• Installing MVAPICH2 GDR
• Split Grid

Using GGP In Your applications

• GGP Interface
• GGP Parameter Structures
• GGP Data Structures And Types
• Best Practices

GGP Development

Contributing to GGP

• Updating the GGP Interface
• Adding new GGP features
• Writing GGP Kernels
• Coding Conventions and Style

Resources

• GGP Development model Some Rules and Best Practices for writing code and use of git
• Continuous Integration
• GGP Release Procedure
• GGP Release Testing
• GGP Debugging
• Timing GGP build
• External GGP mods

Case Studies: Pedagogical Instructions On Using Some of GGP's Key Features

Below we have included some important case studies, designed to be guides on how to get the most out of GGP solvers for different use cases in modern LQCD.

• Tensor Cores