Characterizing the I/O workload and performance of computational science applications has not been as well studied as other aspects of application performance. I/O studies are difficult because applications can use I/O for a variety of different purposes, such as initializing data structures, temporary storage of intermediate results that exceed memory capacity, recording or resuming the […]
Author: [email protected]
Thornado-mini
Thornado-mini is a proxy app developed in the ExaStar ECP AD project. The ExaStar project targets multi-physics simulations of astrophysical explosions. Their target application, Clash, will be a component-based multi-physics toolkit, built on capabilities of current astrophysics simulation codes Flash and Castro and on the massively parallel adaptive mesh refinement framework {AMReX}. The Clash code […]
PICSARlite
PICSARlite is part of the WarpX, a software package for exascale modeling of plasma-based particle accelerators. Its goal is to assist the development of compact and affordable high-energy physics colliders. The components of WarpX are: PICSAR: a library of Fortran 90 subroutines for Particle-In-Cell (PIC) operations. AMReX: a software framework written in C++ and Fortran […]
miniQMC
MiniQMC is a quantum Monte Carlo code that comprises the important computational kernels of its parent, QMCPACK. Quantum Monte Carlo methods are often applied in material science to understand the electronic structure of molecular and solid state systems. The computational motifs of miniQMC and QMCPACK are particle methods, dense and sparse linear algebra, and Monte […]
miniVite
Science MiniVITE is a proxy app for the Vite program. Vite itself is the distributed-memory version of Grappolo. The program performs community detection over a graph, which is used in graph applications to reveal natural divisions that exist in networks without size or element constraints. One obvious example for the application of community detection is […]