Skip to content

Tinker-HP: High-Performance Massively Parallel Evolution of Tinker on CPUs & GPUs

Notifications You must be signed in to change notification settings

TinkerTools/tinker-hp

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Tinker-HP: High-Performance Massively Parallel Evolution of Tinker on CPUs & GPUs

News


  • Update 07/2024: integration of the Lambda-ABF method for alchemical free energy simulations. Check the Lambda-ABF paper (J. Chem. Theory Comput. 024, 20, 11, 4481–4498)
  • Update 08/2023: integration of Quantum-HP for Nuclear Quantum Effects (NQE) through RPMD and (ad)QTB methods. Check the Quantum-HP paper (J. Chem. Theory Comput. 2023, 19, 5, 1432–1445)
  • Update 02/2023: support for neural networks potentials (ANI-2X, DeepMD etc...) is available. Check the Deep-HP module paper (Chem. Sci., 2023,14, 5438-5452)
  • Update 02/2023: support for the AMOEBA+ potential now available
  • Update 10/2022: New website for Tinker-HP, check it out ! https://tinker-hp.org
  • Update 02/2021: PLUMED Support for version 1.2 GPUs
  • Update 11/2021: PLUMED Support for version 1.2 (CPUs)
  • Update 24/2020: all versions have been pushed to GitHub


Versions

Current Github version: 1.1v (enhanced AVX512 vectorized CPUs version), 1.2 (CPUs) + 1.2 (multi)-GPUs
Current Development version: 1.3 (CPUs + multi-GPUs)

All releases of the Tinker-HP code are now being performed on Github. For news, benchmarks and additional tutorials, please visit the Tinker-HP website and follow us on Twitter
In addition to GitHub, a GPUs container (quick install!) is available thanks to NVIDIA on the NVIDIA NGC's website.

Description

Tinker-HP is a CPUs and GPUs based, multi-precision, MPI massively parallel package dedicated to long polarizable molecular dynamics simulations and to polarizable QM/MM. Tinker-HP is an evolution of the popular Tinker package that conserves it simplicity of use but brings new capabilities allowing performing very long molecular dynamics simulations on modern supercomputers that use thousands of cores. The Tinker-HP approach offers various strategies using domain decomposition techniques for periodic boundary conditions in the framework of the (n)log(n) Smooth Particle Mesh Ewald. Tinker-HP proposes a high performance scalable computing environment for polarizable (AMOEBA, Amberpol...) and classical (Amber, Charmm, OPLS...) force fields giving access to large systems up to millions of atoms. It can be used on supercomputers as well as on lab clusters. Tinker-HP supports Intel (AVX-512 enhanced version) and AMD CPUs platforms as well as NVIDIA GPUs (GTX-10xx, RTX-20xx, 30xx, 40xx, P100, V100, A100).

Licence

Tinker-HP is available free of charge for ALL Academic Institutions, National Laboratories and supercomputer centers through the global Tinker license.
Non-academic entities (e.g., companies, for profit organizations) should contact the managing universities (see license).

Please Cite

Tinker-HP is part of the Tinker distribution and uses the same tools as Tinker. These tools can be found here

Contact

We provide support to users:

Email: [email protected]