Engineering Papers⌕ Search

Engineering topics

Ding, Nan

Publications and source records attributed to Ding, Nan.

Evaluating the potential of disaggregated memory systems for HPC applications

Summary Disaggregated memory is a promising approach that addresses the limitations of traditional memory architectures by enabling memory to be decoupled from compute nodes and shared across a data center. Cloud platforms have deployed such systems to improve overall system memory utilization, but performance can vary across workloads. High‐performance computing (HPC) is crucial in scientific and engineering applications, where HPC machines also face the issue of underutilized memory. As a result, improving system memory utilization while understanding workload performance is essential for HPC operators. Therefore, learning the potential of a disaggregated memory system before deployment is a critical step. This paper proposes a methodology for exploring the design space of a disaggregated memory system. It incorporates key metrics that affect performance on disaggregated memory systems: memory capacity, local and remote memory access ratio, injection bandwidth, and bisection bandwidth, providing an intuitive approach to guide machine configurations based on technology trends and workload characteristics. We apply our methodology to analyze thirteen diverse workloads, including AI training, data analysis, genomics, protein, fusion, atomic nuclei, and traditional HPC bookends. Our methodology demonstrates the ability to comprehend the potential and pitfalls of a disaggregated memory system and provides motivation for machine configurations. Our results show that eleven of our thirteen applications can leverage injection bandwidth disaggregated memory without affecting performance, while one pays a rack bisection bandwidth penalty and two pay the system‐wide bisection bandwidth penalty. In addition, we also show that intra‐rack memory disaggregation would meet the application's memory requirement and provide enough remote memory bandwidth.

Ding, Nan↗

Unified Communication Optimization Strategies for Sparse Triangular Solver on CPU and GPU Clusters

This paper presents a unified communication optimization framework for sparse triangular solve (SpTRSV) algorithms on CPU and GPU clusters. The framework builds upon a 3D communication-avoiding (CA) layout of Px × Py × Pz processes that divides a sparse matrix into Pz submatrices, each handled by a Px × Py 2D grid with block-cyclic distribution. We propose three communication optimization strategies: First, a new 3D SpTRSV algorithm is developed, which trades the inter-grid communication and synchronization with replicated computation. This design requires only one inter-grid synchronization, and the inter-grid communication is efficiently implemented with sparse allreduce operations. Second, broadcast and reduction communication trees are used to reduce message latency of the intra-grid 2D communication on CPU clusters. Finally, we leverage GPU-initiated one-sided communication to implement the communication trees on GPU clusters. With these nested inter- and intra-grid communication optimization strategies, the proposed 3D SpTRSV algorithm can attain up to 3.45x speedups compared to the baseline 3D SpTRSV algorithm using up to 2048 Cori Haswell CPU cores. In addition, the proposed GPU 3D SpTRSV algorithm can achieve up to 6.5x speedups compared to the proposed CPU 3D SpTRSV algorithm with Pz up to 64. Finally it is remarkable that the proposed GPU 3D SpTRSV can scale to 256 GPUs using the Perlmutter system while the existing 2D SpTRSV algorithm can only scale up to 4 GPUs.

Sao, Piyush↗

Accelerating large scale de novo metagenome assembly using GPUs

Metagenomic workflows involve studying uncultured microorganisms directly from the environment. These environmental samples when processed by modern sequencing machines yield large and complex datasets that exceed the capabilities of metagenomic software. The increasing sizes and complexities of datasets make a strong case for exascale-capable metagenome assemblers. However, the underlying algorithmic motifs are not well suited for GPUs. This poses a challenge since the majority of next-generation supercomputers will rely primarily on GPUs for computation. In this paper we present the first of its kind GPU-Accelerated implementation of the local assembly approach that is an integral part of a widely used large-scale metagenome assembler, MetaHipMer. Local assembly uses algorithms that induce random memory accesses and non-deterministic workloads, which make GPU offloading a challenging task. Our GPU implementation outperforms the CPU version by about 7x and boosts the performance of MetaHipMer by 42% when running on 64 Summit nodes.

Awan, Muaaz Gul↗

Hardware Evaluation Analytical Modeling and Node Simulation: Benefits of Tighter GPU Integration

In this report, we examine several emerging technologies of interest to the Department of Energy and its computational centers. These include: 1) quantifying the benefit of tighter CPU-GPU integration, 2) quantifying the appropriate CPU core:GPU ratio, 3) quantifying the penalty for CPU-GPU disaggregation, 4) quantifying the benefits of tighter GPU-GPU integration, 5) quantifying the benefits of unified memory, and 6) quantifying the benefits of tighter FPGA-GPU integration.

97 MATHEMATICS AND COMPUTING↗

LOGAN: High-Performance GPU-Based X-Drop Long-Read Alignment

Pairwise sequence alignment is one of the most computationally intensive kernels in genomic data analysis, accounting for more than 90% of the runtime for key bioinformatics applications. This method is particularly expensive for third-generation sequences due to the high computational cost of analyzing sequences of length between 1Kb and 1Mb. Given the quadratic overhead of exact pairwise algorithms for long alignments, the community primarily relies on approximate algorithms that search only for high-quality alignments and stop early when one is not found. In this work, we present the first GPU optimization of the popular X-drop alignment algorithm, that we named LOGAN. Results show that our high-performance multi-GPU implementation achieves up to 181.6 GCUPS and speed-ups up to 6.6× and 30.7× using 1 and 6 NVIDIA Tesla V100, respectively, over the state-of-the-art software running on two IBM Power9 processors using 168 CPU threads, with equivalent accuracy. We also demonstrate a 2.3× LOGAN speed-up versus ksw2, a state-of-art vectorized algorithm for sequence alignment implemented in minimap2, a long-read mapping software. Furthermore, to highlight the impact of our work on a real-world application, we couple LOGAN with a many-to-many long-read alignment software called BELLA, and demonstrate that our implementation improves the overall BELLA runtime by up to 10.6×. Finally, we adapt the Roofline model for LOGAN and demonstrate that our implementation is near optimal on the NVIDIA Tesla V100s.

97 MATHEMATICS AND COMPUTING↗

Organophosphorus esters (OPEs) in PM2.5 in urban and e-waste recycling regions in southern China: concentrations, sources, and emissions

Highlights: • OPE concentrations and compositions displayed great variations among the sources. • High OPE levels were related to the electronic, plastic, and chemical industries. • Correlations between OPEs suggests different emission mechanisms from the sources. • The total air emission of OPEs from the 20 urban industrial sources was 1460 kg/yr. • The total air emission of OPEs from the e-waste recycling facilities was 133 kg/yr. Organophosphate esters (OPEs) are novel ubiquitous contaminants that are attracting growing concern, but their emissions into the environment are still poorly understood. In this study, 12 OPEs were measured in fine particulate matter (PM{sub 2.5}) at 20 industrial sites in an urban region and four e-waste recycling facilities in a rural region in southern China. There was no significant difference in the concentrations of ∑OPEs between the urban region (519–62,747 pg/m{sup 3}, median = 2854 pg/m{sup 3}) and the rural e-waste region (775–13,823 pg/m{sup 3}, 3321 pg/m{sup 3}). High OPE concentrations in urban PM{sub 2.5} were generally associated with the electrical, electronic, plastic, and chemical industries. There were no significant correlations between most OPEs in these two regions, suggesting different emission mechanisms. The average emissions of ∑OPEs estimated using a simplified dispersion model were 73.0 kg/yr from the urban industrial point sources and 33.2 kg/yr from the e-waste recycling facilities. The estimated emission inventory from industrial activities in the whole city (3228–4452 kg/yr) was approximately 30-fold higher than that from the e-waste recycling (133 kg/yr) facilities because urban region has a much larger industrial scale. To the best of our knowledge, this is the first effort to model the emissions of OPEs from industrial and e-waste recycling activities to the atmosphere.

54 ENVIRONMENTAL SCIENCES↗

Characterizing quantum supremacy in near-term devices

A critical question for quantum computing in the near future is whether quantum devices without error correction can perform a well-defined computational task beyond the capabilities of supercomputers. Such a demonstration of what is referred to as quantum supremacy requires a reliable evaluation of the resources required to solve tasks with classical approaches. Here, we propose the task of sampling from the output distribution of random quantum circuits as a demonstration of quantum supremacy. We extend previous results in computational complexity to argue that this sampling task must take exponential time in a classical computer. We introduce cross-entropy benchmarking to obtain the experimental fidelity of complex multiqubit dynamics. This can be estimated and extrapolated to give a success metric for a quantum supremacy demonstration. We study the computational cost of relevant classical algorithms and conclude that quantum supremacy can be achieved with circuits in a two-dimensional lattice of 7 × 7 qubits and around 40 clock cycles. This requires an error rate of around 0.5% for two-qubit gates (0.05% for one-qubit gates), and it would demonstrate the basic building blocks for a fault-tolerant quantum computer.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

An automatic performance model-based scheduling tool for coupled climate system models

The prediction of the climate system is highly depended on the efficient integration of observations and simulations of the Earth. This is regarded as a canonical example of the cyber-physical system. The climate system model, the simulation engine in this cyber-physical system, is one of most challenging applications in scientific computing. It utilizes the multi-physics simulation that couples multiple components, conducts decadal to millennium simulations, and has long been an important application on supercomputers. However, current climate system models suffer from the inefficient task scheduling methods resulting in an intolerable simulation time. Take the Community Earth System Model (CESM), the most widely used climate system model, as an example, one major reason that CESM suffers from bad performances is the huge overhead to rationally distribute processes among the coupled heterogeneous components. According to the report of NCAR, every percent improvement in CESM performance frees up to the equivalent of $250,000 in computing resources in their scientific experiments. To address such challenge, our paper firstly constructs a lightweight and accurate performance model for effectively capturing and predicting the heterogeneous time-to-solution performance of end-to-end CESM components with a given simulation configuration. Then, based on the performance model, we further propose an efficient scheduling strategy based on rectangular packing method to determine the best process layout among different components, and the process numbers assigned to each component. Our evaluations show that we can achieve 58% average run time reductions on CESM comparing to the widely used sequential process layout for a scale of 144 to 480 cores on typical CPU clusters. And we can save 4 million CPU hours when we conduct one standard scientific experiment (a 2870-year simulation), which equals to save $40,089 with a charge of $0.01 per CPU hour. Meanwhile, 26% extra performance improvements could also be gained in our methods comparing to the heuristic branch and bound algorithm with the guidance of the known curve-fitting performance model.

58 GEOSCIENCES↗

MAGE-A1 promotes melanoma proliferation and migration through C-JUN activation

MAGE-A1 belongs to the chromosome X-clustered genes of cancer-testis antigen family and is normally expressed in the human germ line but is also overexpressed in various tumors. Previous studies of MAGE-A1 in melanoma mainly focused on methylation changes or its role in immunotherapy, however, its biological functions in melanoma have remained unknown. In order to determine the role of MAGE-A1 in melanoma growth and metastasis, we manipulated melanoma cell lines with overexpression and knockdown of MAGE-A1. Integration of cell proliferation assays, transwell migration and invasion assays, and RNA-Seq analysis revealed that up-regulation of MAGE-A1 dramatically promoted proliferation, migration, and invasion of human melanoma cell lines in vitro, while down-regulation of MAGE-A1 inhibited those characteristics associated with tumor cells. Furthermore, transcriptome sequencing revealed that MAGE-A1 exerts its tumor promoting activity by activating p-C-JUN directly or through ERK-MAPK signaling pathways. Based on our findings, we propose that MAGE-A1 may be a potential therapeutic target for melanoma patients. - Highlights: • MAGE-A1 promotes proliferation and clone formation in melanoma cell lines. • MAGE-A1 enhances tumor cell migration and invasion in melanoma cell lines. • Network including C-JUN, IL8, and ARHGAP29 play critical role in malignant melanoma. • Oncogenic MAGE-A1 increases p-C-JUN levels, possibly via ERK-MAPK signaling pathway.

60 APPLIED LIFE SCIENCES↗

Facile synthesis of size-tunable CuO/graphene composites and their high photocatalytic performance

Graphical abstract: A novel leaf-like CuO/graphene nanosheet (GNS) with tunable size was prepared by a facile hydrothermal process, and the stirring temperature in synthesis was found to be effective for controlling its size. - Highlights: • Leaf-like CuO/GNS nanocomposites with tunable size was prepared via hydrothermal method. • With increase of stirring temperature, the surface area becomes larger. • The morphology of nanocomposites plays an important role in photocatalysis. - Abstract: Size-tunable leaf-like copper oxide modified by graphene nanosheets was synthesized by a facile hydrothermal method. Stirring temperature control during synthesis was found to be important in effectively controlling the size of CuO nanoparticles, with smaller size CuO nanoparticles produced by increasing the stirring temperature. The as-prepared nanocomposites were characterized by XRD, TEM, and nitrogen adsorption–desorption. The nanocomposites exhibited better catalytic activity than pure CuO toward the degradation of rhodamine B in the presence of H{sub 2}O{sub 2} under visible light irradiation. The catalytic performance was improved with decreasing particle size of the nanocomposites, which was attributed to the increased in the rate of interelectron transfer at the interface.

36 MATERIALS SCIENCE↗