README

This is the README file for aCG, a suite of GPU-accelerated iterative
linear solvers based on the conjugate gradient (CG) method, supporting
NVIDIA and AMD GPUs, as well as multi-GPU systems with GPU-aware MPI,
NCCL, RCCL or NVSHMEM.

For NVIDIA GPUs, CUDA implementations of CG and pipelined CG are
provided. Communication is performed by the host CPU using GPU-aware
MPI, NCCL or NVSHMEM, or by the GPU using NVSHMEM device-initiated
communication.

For AMD GPUs, HIP implementations of CG and pipelined CG are provided.
Communication is performed by the host CPU with GPU-aware MPI or RCCL.
A single-GPU HIP version of a monolithic, device-side CG solver is
also provided.

See the file INSTALL for instructions on how to build and install.

aCG is free software, available under a permissive software license.
See the file LICENSE for copying conditions.

--
Copyright 2025 Koç University and Simula Research Laboratory

Copying and distribution of this file, with or without modification,
are permitted provided the copyright notice and this notice are
preserved.