High-Performance Ptychographic Reconstruction with Federated Facilities

被引:1
|
作者
Bicer, Tekin [1 ,3 ]
Yu, Xiaodong [1 ]
Ching, Daniel J. [3 ]
Chard, Ryan [1 ]
Cherukara, Mathew J. [3 ]
Nicolae, Bogdan [2 ]
Kettimuthu, Rajkumar [1 ]
Foster, Ian T. [1 ]
机构
[1] Argonne Natl Lab, Data Sci & Learning Div, CELS, Lemont, IL 60439 USA
[2] Argonne Natl Lab, Math & Comp Sci Div, CELS, Lemont, IL 60439 USA
[3] Argonne Natl Lab, Xray Sci Div, APS, Lemont, IL 60439 USA
关键词
Ptychography; High-performance computing; Synchrotron light source; Scientific computing; Federation; ALTERNATING DIRECTION METHOD; RAY; FLUORESCENCE; TOMOGRAPHY; MICROSCOPY; WORKFLOWS;
D O I
10.1007/978-3-030-96498-6_10
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
Beamlines at synchrotron light source facilities are powerful scientific instruments used to image samples and observe phenomena at high spatial and temporal resolutions. Typically, these facilities are equipped only with modest compute resources for the analysis of generated experimental datasets. However, high data rate experiments can easily generate data in volumes that take days (or even weeks) to process on those local resources. To address this challenge, we present a system that unifies leadership computing and experimental facilities by enabling the automated establishment of data analysis pipelines that extend from edge data acquisition systems at synchrotron beamlines to remote computing facilities; under the covers, our system uses Globus Auth authentication to minimize user interaction, funcX to run user-defined functions on supercomputers, and Globus Flows to define and execute workflows. We describe the application of this system to ptychography, an ultrahigh-resolution coherent diffraction imaging technique that can produce 100s of gigabytes to terabytes in a single experiment. When deployed on the DGX A100 ThetaGPU cluster at the Argonne Leadership Computing Facility and a microscopy beamline at the Advanced Photon Source, our system performs analysis as an experiment progresses to provide timely feedback.
引用
收藏
页码:173 / 189
页数:17
相关论文
共 50 条
  • [41] High-performance reconstruction of microscopic force fields from Brownian trajectories
    Laura Pérez García
    Jaime Donlucas Pérez
    Giorgio Volpe
    Alejandro V. Arzola
    Giovanni Volpe
    Nature Communications, 9
  • [42] High-Performance Multi-Mode Ptychography Reconstruction on Distributed GPUs
    Dong, Zhihua
    Fang, Yao-Lung L.
    Huang, Xiaojing
    Yan, Hanfei
    Ha, Sungsoo
    Xu, Wei
    Chu, Yong S.
    Campbell, Stuart, I
    Lin, Meifeng
    2018 NEW YORK SCIENTIFIC DATA SUMMIT (NYSDS), 2018,
  • [43] High-Performance Computing Based Approach for Improving Semantic-Based Federated Data Processing
    Algosaibi, Abdulelah A.
    INTERNATIONAL JOURNAL OF MATHEMATICS AND COMPUTER SCIENCE, 2021, 16 (01): : 287 - 309
  • [44] SecureBoost plus : Large Scale and High-Performance Vertical Federated Gradient Boosting Decision Tree
    Fan, Tao
    Chen, Weijing
    Ma, Guoqiang
    Kang, Yan
    Fan, Lixin
    Yang, Qiang
    ADVANCES IN KNOWLEDGE DISCOVERY AND DATA MINING, PT III, PAKDD 2024, 2024, 14647 : 237 - 249
  • [45] HIGH-PERFORMANCE CARS DEMAND HIGH-PERFORMANCE MATERIALS
    MISKA, KH
    MATERIALS ENGINEERING, 1974, 79 (05): : 22 - 24
  • [46] HQsFL: A Novel Training Strategy for Constructing High-performance and Quantum-safe Federated Learning
    Yu, Bo
    Shen, Huajie
    Xu, Qian
    He, Wei
    Mao, Wankui
    Zhang, Qing
    Zhang, Fan
    PROCEEDINGS OF THE 19TH ACM ASIA CONFERENCE ON COMPUTER AND COMMUNICATIONS SECURITY, ACM ASIACCS 2024, 2024, : 512 - 521
  • [47] WHICH HIGH-PERFORMANCE MATERIAL FOR HIGH-PERFORMANCE FASTENING
    ROACH, TA
    MATERIALS ENGINEERING, 1981, 94 (01): : 60 - 64
  • [48] Optimal User Selection for High-Performance and Stabilized Energy-Efficient Federated Learning Platforms
    Jeon, Joohyung
    Park, Soohyun
    Choi, Minseok
    Kim, Joongheon
    Kwon, Young-Bin
    Cho, Sungrae
    ELECTRONICS, 2020, 9 (09) : 1 - 17
  • [49] A High-Performance Federated Learning Aggregation Algorithm Based on Learning Rate Adjustment and Client Sampling
    Gao, Yulian
    Lu, Gehao
    Gao, Jimei
    Li, Jinggang
    MATHEMATICS, 2023, 11 (20)
  • [50] FLASH: Towards a High-performance Hardware Acceleration Architecture for Cross-silo Federated Learning
    Zhang, Junxue
    Cheng, Xiaodian
    Wang, Wei
    Yang, Liu
    Hu, Jinbin
    Chen, Kai
    PROCEEDINGS OF THE 20TH USENIX SYMPOSIUM ON NETWORKED SYSTEMS DESIGN AND IMPLEMENTATION, NSDI 2023, 2023, : 1057 - 1079