Search NASA⌕ Search

Engineering topics

Peng, Zhen

Publications and source records attributed to Peng, Zhen.

Automatic Code Generation for High-Performance Graph Algorithms

Graph problems are common across fields of scientific computing and social sciences. However, despite their importance, implementing graph algorithms effectively on modern computing systems is a challenging task that requires significant programming effort and generally results in customized implementations. Current computing and memory hierarchies are not architected for irregular computations resulting in challenges for graph algorithms to achieve high performance on those architectures. In this paper, we present GraphX, a novel compiler framework and DSL designed to simplify the development of efficient graph algorithms and achieve high performance on modern computing systems. GraphX consists of a DSL for efficient implementation of graph algorithms, various optimizations, such as support for sparse linear algebra and workspace transformations, optimized graph primitives, including semiring and masking, and a high-performance code generation engine. Using GraphX, users can implement graph algorithms using a semantically-rich language with graph-oriented operators. GraphX uses these semantics to automatically generate efficient code for target architectures, increasing performance and portability across architectures. The composable nature of GraphX makes it possible to extend the set of optimizations and architectures without modifying the source code. We demonstrate GraphX outperforms state-of-the-art graph libraries, such as LAGraph, up to $3.7 speedup in semiring operations, $2.19 speedup in an important sparse computational kernel, and $9.05 speedup in graph processing algorithms.

compiler, graph algorithms, semiring, masking, wor↗

In situ study on medium-range order evolution during the polyamorphous phase transition in a Pd-Ni-P nanostructured glass

Engineering multiscale structural hierarchies in glassy alloys enable a broad spectrum of potential applications. Metallic glasses were born in hierarchical structures from atomic-to-nanometer scales. However, the frozen-in structures in traditional metallic glasses prepared by rapid quenching techniques are challenging to tailor. Here, we show that a Pd 40 Ni 40 P 20 bulk nanostructured glass of polyamorphous interfacial structures was prepared by inert-gas condensation with a laser evaporation source, and its multiscale structures could be engineered. In-situ scattering experiment results reveal polyamorphous phase transitions occurred in the interfacial regions, which are accompanied by the evolution of medium-range order and the nanoscale heterogeneous structures during the condensation process of glassy nanoparticles under high pressure and the following heating process. Moreover, changes in the cluster connectivity resulting from repacking of the local ordering induced by pressure and temperature could be observed. The thermophysical and mechanical properties, including boson peaks, hardness, and elasticity modulus, could be changed as a function of heat-treatment parameters. In conclusion, our findings would shed light on the synthesis of bulk nanostructured glassy alloys with tailorable thermodynamic and dynamical behavior as well as mechanical properties based on the understanding of metastability for polyamorphous interfacial phases.

36 MATERIALS SCIENCE↗

MemHC: An Optimized GPU Memory Management Framework for Accelerating Many-body Correlation

The many-body correlation function is a fundamental computation kernel in modern physics computing applications, e.g., Hadron Contractions in Lattice quantum chromodynamics (QCD). This kernel is both computation and memory intensive, involving a series of tensor contractions, and thus usually runs on accelerators like GPUs. Existing optimizations on many-body correlation mainly focus on individual tensor contractions (e.g., cuBLAS libraries and others). In contrast, this work discovers a new optimization dimension for many-body correlation by exploring the optimization opportunities among tensor contractions. More specifically, it targets general GPU architectures (both NVIDIA and AMD) and optimizes many-body correlation’s memory management by exploiting a set of memory allocation and communication redundancy elimination opportunities: first, GPU memory allocation redundancy: the intermediate output frequently occurs as input in the subsequent calculations; second, CPU-GPU communication redundancy: although all tensors are allocated on both CPU and GPU, many of them are used (and reused) on the GPU side only, and thus, many CPU/GPU communications (like that in existing Unified Memory designs) are unnecessary; third, GPU oversubscription: limited GPU memory size causes oversubscription issues, and existing memory management usually results in near-reuse data eviction, thus incurring extra CPU/GPU memory communications.

97 MATHEMATICS AND COMPUTING↗

Unique microporous NbO-type Co{sup II}/Zn{sup II} MOFs from double helical chains: Sorption and luminescent properties

Under solvothermal conditions, the reactions of Co{sup II}/Zn{sup II} ions with bent ligand 3-(pyridin-4-yl)-5-(pyrazin-2-yl)-1H-1,2,4-triazole (4-Hpzpt) afford two compounds {[M(4-pzpt)_2] guest}{sub n} (guest=H{sub 2}O, M=Co{sup II} (1), Zn{sup II} (2)). Both compounds are the thermally and hydrolytically robust 4-connected 3D NbO framework, which formed by double helical chains to give rise to 1D hollow nanochannel with uncoordinated nitrogens completely exposed on the pore surface. Compound 1 exhibits improved N{sub 2}, CO{sub 2} and H{sub 2} uptake capacities, while compound 2 displays the strong luminescent emission with obvious red shift. - Graphical abstract: Two 2-fold interpenetrated NbO-type MOFs with 1D nanochannel were synthesized. Compound 1 exhibits improved N{sub 2}, CO{sub 2} and H{sub 2} uptake capacities, while compound 2 displays the strong fluorescent emission with obvious red shift. Display Omitted.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗