Regent

Abstract

We present Regent, a high-productivity programming language for high performance computing with logical regions. Regent users compose programs with tasks (functions eligible for parallel execution) and logical regions (hierarchical collections of structured objects). Regent programs appear to execute sequentially, require no explicit synchronization, and are trivially deadlock-free. Regent's type system catches many common classes of mistakes and guarantees that a program with correct serial execution produces identical results on parallel and distributed machines. We present an optimizing compiler for Regent that translates Regent programs into efficient implementations for Legion, an asynchronous task-based model. Regent employs several novel compiler optimizations to minimize the dynamic overhead of the runtime system and enable efficient operation. We evaluate Regent on three benchmark applications and demonstrate that Regent achieves performance comparable to hand-tuned Legion.

Evaluation of finite difference based asynchronous partial differential equations solver for reacting flows

“…A similar approach has been utilized in [58] where asynchronous copies between CPUs and GPUs are used to overlap computation and data movement, but delayed data with asynchronous computations have not been used. At a compiler level, new asynchronous run-time systems that are capable of dynamic task parallelism are being developed to improve the computation-communication overlap [59][60][61]. The AT schemes, which relax synchronization at a mathematical level, can be coupled with such data-centric programming models to create highly scalable PDE solvers.…”

A framework for modeling human behavior in large-scale agent-based epidemic simulations

“…PanSim makes heavy use of the nonblocking MPI communication primitives to overlap computation and communication phases. While many other distributed memory HPC messaging frameworks have been developed over the years (such as Charm++, 82 UPC++, 83 and Legion and Regent 84 ), MPI remains the most widely supported HPC framework on HPC platforms.…”

Direct numerical simulation of a turbulent methane/air flame impinged by a sub-breakdown electric field

“…The calculations presented in this work are carried out with the described numerical procedure in the Hypersonic Task-based Research (HTR) solver [56][57][58]. The HTR solver leverages the runtime Legion [59][60][61] and the programming language Regent [62] to execute the described calculations on high-performance supercomputers with heterogeneous architectures. Specifically, the supercomputing facility Lassen at the Lawrence Livermore National Laboratory (LLNL) has been utilized in the present work.…”