The Specialized High-Performance Network on Anton 3

  title={The Specialized High-Performance Network on Anton 3},
  author={Keun Sup Shim and Brian Greskamp and Brian Towles and Bruce Edwards and Jerry P. Grossman and David E. Shaw},
  journal={2022 IEEE International Symposium on High-Performance Computer Architecture (HPCA)},
  • Keun Sup Shim, B. Greskamp, D. Shaw
  • Published 20 January 2022
  • Computer Science
  • 2022 IEEE International Symposium on High-Performance Computer Architecture (HPCA)
Molecular dynamics (MD) simulation, a computationally intensive method that provides invaluable insights into the behavior of biomolecules, typically requires large-scale parallelization. Implementation of fast parallel MD simulation demands both high bandwidth and low latency for inter-node communication, but in current semiconductor technology, neither of these properties is scaling as quickly as intra-node computational capacity. This disparity in scaling necessitates architectural… 


Anton 2: Raising the Bar for Performance and Programmability in a Special-Purpose Molecular Dynamics Supercomputer
  • D. E. Shaw, J. P. Grossman, C. Young
  • Computer Science, Physics
    SC14: International Conference for High Performance Computing, Networking, Storage and Analysis
  • 2014
The architecture of Anton 2 is tailored for fine-grained event-driven operation, which improves performance by increasing the overlap of computation with communication, and also allows a wider range of algorithms to run efficiently, enabling many new software-based optimizations.
Anton, a special-purpose machine for molecular dynamics simulation
A massively parallel machine called Anton is described, which should be capable of executing millisecond-scale classical MD simulations of such biomolecular systems and is designed to use both novel parallel algorithms and special-purpose logic to dramatically accelerate those calculations that dominate the time required for a typical MD simulation.
Anton 3: twenty microseconds of molecular dynamics simulation before lunch
The main architectural and algorithmic developments that were necessary to achieve significant improvements in time-to-solution over its predecessor, Anton 2, and over 100-fold faster than any other currently available supercomputer are presented.
Scalable molecular dynamics on CPU and GPU architectures with NAMD.
The main features of NAMD are reviewed, including the variety of options offered by NAMD for enhanced-sampling simulations aimed at determining free-energy differences of either alchemical or geometrical transformations and their applicability to specific problems.
MDGRAPE-4: a special-purpose computer system for molecular dynamics simulations
The MDGRAPE-4 system will enable long-time molecular dynamics simulations of small systems and is also useful for multiscale molecular simulations where the particle simulation parts often become bottlenecks.
New parallel computing algorithm of molecular dynamics for extremely huge scale biological systems
In this paper, we address high performance extreme‐scale molecular dynamics (MD) algorithm in the GENESIS software to perform cellular‐scale molecular dynamics (MD) simulations with more than 100,000
Heterogeneous Parallelization and Acceleration of Molecular Dynamics Simulations in GROMACS
This work presents the heterogeneous parallelization and acceleration design of molecular dynamics implemented in the GROMACS codebase over the last decade, and enables excellent performance from single GPU simulations through strong scaling across multiple GPUs and efficient multi-node parallelization.
Scalable molecular dynamics with NAMD on the Summit system
Algorithm improvements and performance optimizations that enable NAMD to achieve high performance on the IBM Newell platform (with POWER9 processors and NVIDIA Volta V100 GPUs) which underpins the Oak Ridge National Laboratory's Summit and Lawrence Livermore National Lab's Sierra supercomputers are presented.
Fully integrated FPGA molecular dynamics simulations
This paper presents what it is believed to be the first full-scale FPGA-based simulation engine, and shows that its performance is competitive with a GPU (running Amber in an industrial production environment).
Exploiting 162-Nanosecond End-to-End Communication Latency on Anton
  • R. Dror, J. P. Grossman, D. E. Shaw
  • Computer Science
    2010 ACM/IEEE International Conference for High Performance Computing, Networking, Storage and Analysis
  • 2010
Key elements of Anton's approach, in addition to tightly integrated communication hardware, include formulating data transfer in terms of counted remote writes, leveraging fine-grained communication, and establishing fixed, optimized communication patterns.