NAMD provides high-performance simulation of large biomolecular systems. Based on Charm++ parallel objects, NAMD scales to hundreds of cores for typical simulations and beyond 500,000 cores for the largest simulations.

NAMD

Charades is a Charm++ parallel discrete event simulation framework. It supports both optimistic and conservative synchronization protocols and multiple GVT algorithms, in addition to other features provided by the Charm++ runtime system, such as dynamic load balancing.

Charades

SpECTRE is an open-source code for multi-scale, multi-physics problems in astrophysics and gravitational physics.

SpECTRE

ChaNGa (Charm++ N-Body Gravity Simulator) simulates the formation of galaxies and large scale structures in the universe with gravity and hydrodynamics calculations. ChaNGa uses Charm++’s dynamic load balancing support to obtain good performance on massively parallel systems, and has shown to scale well up to 512,000 processors on NCSA Blue Waters

ChaNGa

Cello is a Charm++ framework for multi-physics adaptive mesh refinement (AMR) simulations. The design of Enzo-P and Cello relies on fully-distributed Charm++ object array data structures.

Enzo-P/Cello

ADHydro is a large-scale, high-resolution, multi-physics watershed simulation being developed by the CI-WATER watershed modeling team. Parallelization was achieved using the CHARM++ run-time environment.

ADHydro

EpiSimdemics is used to simulate the spread of contagion in large, realistic social contact networks using individual agent-based models. EpiSimdemics uses the small message aggregation and quiescence detection features of Charm++ to achieve best-in-class performance for large scale agent based simulation.

EpiSimdemics

OpenAtom is a highly parallel atomistic simulation framework. It features electronic structure simulation scaling to over 10x cores per state and a new excited state computation (GW) with unprecedented scalability. OpenAtom uses Charm++’s topology aware mapping to optimize communication.

OpenAtom

Charm++ was utilized to address the challenges of parallelizing Asynchronous Contact Mechanics (ACM) and has shown good strong scaling of ACM to 384 cores of a Cray XC30, compared to 30 cores of previously published implementations.

Disney Clothsim