Using hybrid parallelism to improve memory use in the Uintah framework

Meng, Qiaoling; Berzins, Martin; Schmidt, John A.

doi:10.1145/2016741.2016767

Cited by 23 publications

(43 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As Uintah variables can be accessed freely by any thread, many shared data structures, such as data warehouse and task queues, were redesigned to guarantee thread-safety. Experimental results [1] on typical fluid AMR simulations showed 50% to 90% savings on memory usage. This new multi-threaded MPI scheduler enabled Uintah to scale up to [5] 196K cores on the DoE Jaguar XT5 system and became the basis for the heterogeneous multi-threaded MPI scheduler [5] which allowed Uintah to dispatch tasks to GPUs as well as CPU cores on a node.…”

Section: B Multi-threaded Cpu Scheduler (Master-slave Model) [1]mentioning

confidence: 97%

“…A new multi-threaded MPI scheduler (Figure 1, from [1]) was designed in [1] to eliminate intra-node MPI messages and memory copies by adopting a shared memory model on-node. This was realized by creating multiple worker threads on the same multi-core node.…”

Section: B Multi-threaded Cpu Scheduler (Master-slave Model) [1]mentioning

confidence: 99%

“…An important trend in high performance computing is the planning and design of software framework architectures for emerging and future systems with multi-petaflop and eventually exaflop performance [1]. Such frameworks must address the formidable scalability and performance challenges associated with running on these systems, and must also insulate application developers from the inherent complexity of the parallelism involved.…”

Section: Introductionmentioning

confidence: 99%

“…A multi-threaded MPI scheduler was then used [1] to solve this problem by dynamically assigning tasks to worker threads during execution. This multi-threaded MPI scheduler used a master-slave model which had one control thread and several worker threads per MPI process.…”

Section: Introductionmentioning

confidence: 99%

“…In this work, we examine the evolution of Uintah's hybrid multi-threaded MPI runtime system [1] to support, schedule and execute both CPU and GPU tasks simultaneously, without a central control thread. In what follows Section 2 provides an overview of the Uintah software, while Section 3 covers the history of Uintah task schedulers and their respective designs and ultimate scalability and performance barriers.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

The uintah framework: a unified heterogeneous task scheduling and runtime system

Meng

Humphrey

Berzins

2012

2012 SC Companion: High Performance Computing, Networking Storage and Analysis

View full text Add to dashboard Cite

The development of a new unified, multi-threaded runtime system for the execution of asynchronous tasks on heterogeneous systems is described in this work. These asynchronous tasks arise from the Uintah framework, which was developed to provide an environment for solving a broad class of fluid-structure interaction problems on structured adaptive grids. Uintah has a clear separation between its MPI-free user-coded tasks and its runtime system that ensures these tasks execute efficiently. This separation also allows for complete isolation of the application developer from the complexities involved with the parallelism Uintah provides. While we have designed scalable runtime systems for large CPU core counts, the emergence of heterogeneous systems, with additional on-node accelerators and co-processors presents additional design challenges in terms of effectively utilizing all computational resources on-node and managing multiple levels of parallelism. Our work addresses these challenges for Uintah by the development of new hybrid runtime system and Unified multi-threaded MPI task scheduler, enabling Uintah to fully exploit current and emerging architectures with support for asynchronous, out-of-order scheduling of both CPU and GPU computational tasks. This design coupled with an approach that uses MPI to communicate between nodes, a shared memory model on-node and the use of novel lock-free data structures, has made it possible for Uintah to achieve excellent scalability for challenging fluid-structure problems using adaptive mesh refinement on as many as 256K cores on the DoE Jaguar XK6 system. This design has also demonstrated an ability to run capability jobs on the heterogeneous systems, Keeneland and TitanDev. In this work, the evolution of Uintah and its runtime system is examined in the context of our new Unified multi-threaded scheduler design. The performance of the Unified scheduler is also tested against previous Uintah scheduler and runtime designs over a range of processor core and GPU counts.

show abstract

Section: B Multi-threaded Cpu Scheduler (Master-slave Model) [1]mentioning

confidence: 97%

Section: B Multi-threaded Cpu Scheduler (Master-slave Model) [1]mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

The uintah framework: a unified heterogeneous task scheduling and runtime system

Meng

Humphrey

Berzins

2012

2012 SC Companion: High Performance Computing, Networking Storage and Analysis

View full text Add to dashboard Cite

show abstract

Scalable large‐scale fluid–structure interaction solvers in the Uintah framework via hybrid task‐based parallelism algorithms

Meng

Berzins

2013

Concurrency and Computation

Self Cite

View full text Add to dashboard Cite

Uintah is a software framework that provides an environment for solving fluid-structure interaction problems on structured adaptive grids for large-scale science and engineering problems involving the solution of partial differential equations. Uintah uses a combination of fluid flow solvers and particle-based methods for solids, together with adaptive meshing and a novel asynchronous task-based approach with fully automated load balancing. When applying Uintah to fluid-structure interaction problems, the combination of adaptive meshing and the movement of structures through space present a formidable challenge in terms of achieving scalability on large-scale parallel computers. The Uintah approach to the growth of the number of core counts per socket together with the prospect of less memory per core is to adopt a model that uses MPI to communicate between nodes and a shared memory model on-node so as to achieve scalability on large-scale systems. For this approach to be successful, it is necessary to design data structures that large numbers of cores can simultaneously access without contention. This scalability challenge is addressed here for Uintah, by the development of new hybrid runtime and scheduling algorithms combined with novel lock-free data structures, making it possible for Uintah to achieve excellent scalability for a challenging fluid-structure problem with mesh refinement on as many as 260K cores. ‡ Titan is a parallel computer under construction at Oak Ridge National Laboratory with about 299K CPU cores now with a large number of attached GPUs to be added in 2012. § Stampede is a parallel computer under construction at Texas Advanced Computing Center with two petaflops of CPU performance and eight petaflops of accelerator performance. 1389This work will start to address some of these challenges by developing scheduling algorithms and associated software that will be used to tackle fluid-structure interaction algorithms in which mesh refinement is used to resolve the structure. The vehicle for this work is the Uintah Computational Framework [2][3][4]. The broad range and challenging nature of Uintah applications and the software itself are described in [5]. The Uintah code was designed to solve fluid-structure interaction problems arising from a number of physical scenarios. Uintah uses a combination of computational fluid dynamics (CFD) algorithms in its implicit continuous-fluid Eulerian (ICE) solver [6,7] and couples ICE to a particle-based solver for solids known as the material point method (MPM) [8,9]. The adaptive mesh refinement (AMR) approach used in Uintah is that of multiple levels of regularly refined mesh patches. This mesh is also used as a scratch pad for the MPM calculation of the movement and deformation of the solid [10,11]. Uintah has been shown to scale successfully with many fluid and solid problems with adaptive meshes [12,13]; however, the scalability of fluid-structure interaction problems has proven to be somewhat more challenging. This is at least partly because the part...

show abstract

Packing Configurations of PBX‐9501 Cylinders to Reduce the Probability of a Deflagration to Detonation Transition (DDT)

Beckvermit

Harman

Wight

et al. 2016

Propellants Explo Pyrotec

Self Cite

View full text Add to dashboard Cite

The detonation of hundreds of explosive devices from either a transportation or storage accident is an extremely dangerous event. This paper focuses on identifying ways of packing/storing arrays of explosive cylinders that will reduce the probability of a Deflagration to Detonation Transition (DDT). The Uintah Computational Framework was utilized to predict the conditions necessary for a large scale DDT to occur. The results showed that the arrangement of the explosive cylinders and the number of devices packed in a “box” greatly effects the probability of a detonation.

show abstract

Using hybrid parallelism to improve memory use in the Uintah framework

Cited by 23 publications

References 22 publications

The uintah framework: a unified heterogeneous task scheduling and runtime system

The uintah framework: a unified heterogeneous task scheduling and runtime system

Scalable large‐scale fluid–structure interaction solvers in the Uintah framework via hybrid task‐based parallelism algorithms

Packing Configurations of PBX‐9501 Cylinders to Reduce the Probability of a Deflagration to Detonation Transition (DDT)

Contact Info

Product

Resources

About