Christian Robert Trott

Spectral neighbor analysis method for automated generation of quantum-accurate interatomic potentials

We present a new interatomic potential for solids and liquids called Spectral Neighbor Analysis Potential (SNAP). The SNAP potential has a very general form and uses machine-learning techniques to reproduce the energies, forces, and stress tensors of a large set of small configurations of atoms, which are obtained using high-accuracy quantum electronic structure (QM) calculations. The local environment of each atom is characterized by a set of bispectrum components of the local neighbor density projected onto a basis of hyperspherical harmonics in four dimensions. The bispectrum components are the same bond-orientational order parameters employed by the GAP potential 1]. The SNAP potential, unlike GAP, assumes a linear relationship between atom energy and bispectrum components. The linear SNAP coefficients are determined using weighted least-squares linear regression against the full QM training set. This allows the SNAP potential to be fit in a robust, automated manner to large QM data sets using many bispectrum components. The calculation of the bispectrum components and the SNAP potential are implemented in the LAMMPS parallel molecular dynamics code. We demonstrate that a previously unnoticed symmetry property can be exploited to reduce the computational cost of the force calculations by more than one order of magnitude. We present results for a SNAP potential for tantalum, showing that it accurately reproduces a range of commonly calculated properties of both the crystalline solid and the liquid phases. In addition, unlike simpler existing potentials, SNAP correctly predicts the energy barrier for screw dislocation migration in BCC tantalum.

Performance-Portable Sparse Matrix-Matrix Multiplication for Many-Core Architectures

We consider the problem of writing performance portablesparse matrix-sparse matrix multiplication (SPGEMM) kernelfor many-core architectures. We approach the SPGEMMkernel from the perspectives of algorithm design and implementation, and its practical usage. First, we design ahierarchical, memory-efficient SPGEMM algorithm. We thendesign and implement thread scalable data structures thatenable us to develop a portable SPGEMM implementation. We show that the method achieves performance portabilityon massively threaded architectures, namely Intels KnightsLanding processors (KNLs) and NVIDIAs Graphic ProcessingUnits (GPUs), by comparing its performance to specializedimplementations. Second, we study an important aspectof SPGEMMs usage in practice by reusing the structure ofinput matrices, and show speedups up to 3× compared to thebest specialized implementation on KNLs. We demonstratethat the portable method outperforms 4 native methods on2 different GPU architectures (up to 17× speedup), and it ishighly thread scalable on KNLs, in which it obtains 101× speedup on 256 threads.

SNAP: Strong Scaling High Fidelity Molecular Dynamics Simulations on Leadership-Class Computing Platforms

The rapidly improving compute capability of contemporary processors and accelerators is providing the opportunity for significant increases in the accuracy and fidelity of scientific calculations. In this paper we present performance studies of a new molecular dynamics MD potential called SNAP. The SNAP potential has shown great promise in accurately reproducing physics and chemistry not described by simpler potentials. We have developed new algorithms to exploit high single-node concurrency provided by three different classes of machine: the Titan GPU-based system operated by Oak Ridge National Laboratory, the combined Sequoia and Vulcan BlueGene/Q machines located at Lawrence Livermore National Laboratory, and the large-scale Intel Sandy Bridge system, Chama, located at Sandia. Our analysis focuses on strong scaling experiments with approximately 246,000 atoms over the range 1-122,880 nodes on Sequoia/Vulcan and 40-18,630 nodes on Titan. We compare these machine in terms of both simulation rate and power efficiency. We find that node performance correlates with power consumption across the range of machines, except for the case of extreme strong scaling, where more powerful compute nodes show greater efficiency. This study is a unique assessment of a challenging, scientifically relevant calculation running on several of the worlds leading contemporary production supercomputing platforms.

Evaluating the feasibility of using memory content similarity to improve system resilience

Building the next-generation of extreme-scale distributed systems will require overcoming several challenges related to system resilience. As the number of processors in these systems grows, the failure rate increases proportionally. One of the most common sources of failure in large-scale systems is memory errors. In this paper, we propose a novel run-time for transparently exploiting memory content similarity to improve system resilience by reducing the rate at which memory errors lead to node failure. We evaluate the feasibility of this approach by examining memory snapshots collected from eight HPC applications. Based on the characteristics of the similarity that we uncover in these applications, we conclude that our proposed approach shows promise for addressing system resilience in large-scale systems.

An Examination of Content Similarity within the Memory of HPC Applications

Memory content similarity has been e ectively exploited for more than a decade to reduce memory consumption. By consolidating duplicate and similar pages in the address space of an application, we can reduce the amount of memory it consumes without negatively a ecting the applications perception of the memory resources available to it. In addition to memory de-duplication, there may be many other ways that we can exploit memory content similarity to improve system characteristics. In this paper, we examine the memory content similarity of several HPC applications. By characterizing the memory contents of these applications, we hope to provide a basis for ef- forts to e ectively exploit memory content similarity to improve system performance beyond memory deduplication. We show that several applications exhibit signi cant similarity and consider the source of the similarity.

A study of the viability of exploiting memory content similarity to improve resilience to memory errors

Building the next-generation of extreme-scale distributed systems will require overcoming several challenges related to system resilience. As the number of processors in these systems grow, the failure rate increases proportionally. One of the most common sources of failure in large-scale systems is memory. In this paper, we propose a novel runtime for transparently exploiting memory content similarity to improve system resilience by reducing the rate at which memory errors lead to node failure. We evaluate the viability of this approach by examining memory snapshots collected from eight high-performance computing (HPC) applications and two important HPC operating systems. Based on the characteristics of the similarity uncovered, we conclude that our proposed approach shows promise for addressing system resilience in large-scale systems.

Multithreaded sparse matrix-matrix multiplication for many-core and GPU architectures

Sparse Matrix-Matrix multiplication is a key kernel that has applications in several domains such as scientific computing and graph analysis. Several algorithms have been studied in the past for this foundational kernel. In this paper, we develop parallel algorithms for sparse matrix-matrix multiplication with a focus on performance portability across different high performance computing architectures. The performance of these algorithms depend on the data structures used in them. We compare different types of accumulators in these algorithms and demonstrate the performance difference between these data structures. Furthermore, we develop a meta-algorithm, kkSpGEMM, to choose the right algorithm and data structure based on the characteristics of the problem. We show performance comparisons on three architectures and demonstrate the need for the community to develop two phase sparse matrix-matrix multiplication implementations for efficient reuse of the data structures involved.