RADICAL-Pilot (RP) is a flexible Pilot System that provides a simple and scalable approach for executing many concurrent simulations clusters and grids. RP is written in Python and is deployable into user space. It allows user-level control of HPC resources and supports a wide range of application types. It is built on top of the Simple API for Grid Applications (RADICAL-SAGA), a high-level, easy-to-use API for accessing distributed resources. RP thus works on a variety of backends such as PBS, LSF, Slurm etc.
Join the mailinglist for information on releases and supported use cases.
Scalable: Capable of scheduling O(10k) tasks and marshalling O(100k) distributed cores, RADICAL-Pilot offers its users a lightweight API capable of handling a variety of workloads.
Versatile: The sky’s the limit as far as applications are concerned. With use cases ranging from molecular dynamics to astrophysics, many projects can benefit from using RADICAL-Pilot.
Easy-to-use: RADICAL-Pilot allows its users to submit tasks to a variety of HPC machines using only a single, straightforward API. It is deployed in user space, and does not system support nor additional permissions.
Fast: Submit all your tasks via only one job using RADICAL-Pilot, avoiding unnecessary batch scheduling times and significantly reducing time-to-completion.
RP is being used by Chemistry researchers to support large-scale and multidimensional replica exchange simulations on supercomputers.
Researchers at UCL London are using RP to advance understanding of HIV drug resistance and make personalized treatment possible.
RP supports multi-physics and coupled simulations, such as hybrid CFD-MD simulations to understand Couette Flow, as well as PBM-DEM simulations for Cybermanufacturing.