Large‐Scale Plane‐Wave‐Based Density Functional Theory: Formalism, Parallelization, and Applications

Bylaska, Eric J.; Tsemekhman, Kiril; Govind, Niranjan; Valiev, Marat

doi:10.1002/9780470930779.ch3

Cited by 39 publications

(46 citation statements)

References 121 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Small matrix products arise, for example, in electronic structure calculations (e.g. ab-initio molecular dynamics using planewave bases [42,14]). At a high level, the 2.5D algorithm generalizes the traditional 2D Cannon algorithm [50] by employing an additional process dimension to replicate the 2D process grid.…”

Section: 5d Cannon's Algorithmmentioning

confidence: 99%

“…The core layer transforms a minimal subset of MPI point-to-point routines (Bamboo does not support MPI's one sided communication), whereas the second translates a subset of higher level MPI functionality into equivalent point-to-point encodings, which will be then translated by the core layer. This multi-layer design allows one to customize MPI collectives, which may benefit from a specialized interconnect topology or be tailored to the application [14].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Automatic translation of MPI source into a latency-tolerant, data-driven form

Nguyen

Cicotti

Bylaska

et al. 2017

Journal of Parallel and Distributed Computing

Self Cite

View full text Add to dashboard Cite

Hiding communication behind useful computation is an important performance programming technique but remains an inscrutable programming exercise even for the expert. We present Bamboo, a code transformation framework that can realize communication overlap in applications written in MPI without the need to intrusively modify the source code. We reformulate MPI source into a task dependency graph representation, which partially orders the tasks, enabling the program to execute in a data-driven fashion under the control of an external runtime system. Experimental results demonstrate that Bamboo significantly reduces communication delays while requiring only modest amounts of programmer annotation for a variety of applications and platforms, including those employing co-processors and accelerators. Moreover, Bamboo's performance meets or exceeds that of labor-intensive hand coding. The translator is more than a means of hiding communication costs automatically; it demonstrates the utility of semantic level optimization against a well-known library.

show abstract

Section: 5d Cannon's Algorithmmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Automatic translation of MPI source into a latency-tolerant, data-driven form

Nguyen

Cicotti

Bylaska

et al. 2017

Journal of Parallel and Distributed Computing

Self Cite

View full text Add to dashboard Cite

show abstract

“…The 2.5D "communication avoiding" variant of matrix multiplication algorithm [4] targets small matrices. These small matrices arise in the structure calculations targeted by this project [5], [6]. The 2.5D algorithm is interesting for two reasons.…”

Section: 5dmentioning

confidence: 99%

“…This work was featured on the [9] of the Journal of Computational Chemistry [8]. This work was also featured in a cover article and a book chapter [10]. These covers are shown in Fig.…”

Section: New Algorithms For Ab--initio Molecular Dynamicsmentioning

confidence: 99%

Automatic Transformation of MPI Programs to Asynchronous, Graph-Driven Form

Baden¹,

Weare²,

Bylaska³

2013

View full text Add to dashboard Cite

The goals of this project are to develop new, scalable, high-fidelity algorithms for atomic-level simulations and program transformations that automatically restructure existing applications, enabling them to scale forward to Petascale systems and beyond. The techniques enable legacy MPI application code to exploit greater parallelism though increased latency hiding and improved workload assignment. The techniques were successfully demonstrated on high-end scalable systems located at DOE laboratories. Besides the automatic MPI program transformations efforts, the project also developed several new scalable algorithms for ab-initio molecular dynamics, including new massively parallel algorithms for hybrid DFT and new parallel in time algorithms for molecular dynamics and ab-initio molecular dynamics. These algorithms were shown to scale to very large number of cores, and they were designed to work in the latency hiding framework developed in this project. The effectiveness of the developments was enhanced by the direct application to real grand challenge simulation problems covering a wide range of technologically important applications, time scales and accuracies. These included the simulation of the electronic structure of mineral/fluid interfaces, the very accurate simulation of chemical reactions in microsolvated environments, and the simulation of chemical behavior in very large enzyme reactions.

show abstract

“…Small matrix products arise, for example, in electronic structure calculations (e.g. ab-initio molecular dynamics using planewave bases [25], [8]), a planned target of Bamboo. The 2.5D algorithm is interesting for two reasons.…”

Section: Experimental Evaluation a Applications And Testbedmentioning

confidence: 99%

Bamboo -- Translating MPI applications to a latency-tolerant, data-driven form

Nguyen¹,

Cicotti²,

Bylaska³

et al. 2012

2012 International Conference for High Performance Computing, Networking, Storage and Analysis

Self Cite

View full text Add to dashboard Cite

Abstract-We present Bamboo, a custom source-to-source translator that transforms MPI C source into a data-driven form that automatically overlaps communication with available computation. Running on up to 98304 processors of NERSC's Hopper system, we observe that Bamboo's overlap capability speeds up MPI implementations of a 3D Jacobi iterative solver and Cannon's matrix multiplication. Bamboo's generated code meets or exceeds the performance of hand optimized MPI, which includes split-phase coding, the method classically employed to hide communication. We achieved our results with only modest amounts of programmer annotation and no intrusive reprogramming of the original application source.

show abstract

Large‐Scale Plane‐Wave‐Based Density Functional Theory: Formalism, Parallelization, and Applications

Cited by 39 publications

References 121 publications

Automatic translation of MPI source into a latency-tolerant, data-driven form

Automatic translation of MPI source into a latency-tolerant, data-driven form

Automatic Transformation of MPI Programs to Asynchronous, Graph-Driven Form

Bamboo -- Translating MPI applications to a latency-tolerant, data-driven form

Contact Info

Product

Resources

About