The Pacific Research Platform: The First Six Years
Download as PPTX, PDF0 likes116 views
The document describes the achievements and future plans of the Pacific Research Platform (PRP) during its first six years, highlighting its role in connecting science DMZs to support a community-driven cyberinfrastructure. It details the installation of high-speed data transfer nodes, the integration of machine learning capabilities, and collaborative efforts across multiple UC campuses. Key applications focus on addressing critical areas such as climate change, biological data generation, and advanced simulations.
![Top 20 GPU Users Out of 400 Nautilus Namespace Applications:
Together They Consumed Nearly 500 GPUs in 2020
Frank Wuerthwein, UCSD
osggpus [IceCube]
Mark Alber, UCR
markalbergroup
Nuno Vasconcelos, UCSD
domain-adaptation
Ravi Ramamoorthi, UCSD
ucsd-ravigroup
Hao Su, UCSD
ucsd-haosulab
Folding@Home
folding
Igor Sfiligoi, UCSD
isfiligoi
Xiaolong Wang, UCSD
rl-multitask
Xiaolong Wang, UCSD
rl-multitask
Xiaolong Wang, UCSD
self-supervised-video
Xiaolong Wang, UCSD
hand-object-interaction
Dinesh Bharadia, UCSD
ecepxie
Manmohan Chandraker, UCSD
mc-lab
Frank Wuerthwein, UCSD
cms-ml
Nuno Vasconcelos, UCSD
svcl-oowl
Vineet Bafna, UCSD
ecdna
Larry Smarr, UCSD
jupyterlab
Rose Yu, UCSD
deep-forecast
Nuno Vasconcelos, UCSD
svcl-multimodal-learning
Gary Cottrell, UCSD
guru-research](https://image.slidesharecdn.com/prpcapstonesymposiumjune2021-240725184208-9c3d95e6/85/The-Pacific-Research-Platform-The-First-Six-Years-13-320.jpg)
The Pacific Research Platform: The First Six Years
- 1. “The Pacific Research Platform The First Six Years” PRP Capstone Symposium Virtual Meeting June 22, 2021 1 Dr. Larry Smarr Founding Director Emeritus, California Institute for Telecommunications and Information Technology; Distinguished Professor Emeritus, Dept. of Computer Science and Engineering Jacobs School of Engineering, UCSD http://lsmarr.calit2.net
- 2. (GDC) 2015 Vision: The Pacific Research Platform Will Connect Science DMZs Creating a Regional End-to-End Science-Driven Community Cyberinfrastructure NSF CC*DNI Grant $6.3M 10/2015-10/2020 In Year 5 Now PI: Larry Smarr, UC San Diego Calit2 Co-PIs: • Camille Crittenden, UC Berkeley CITRIS, • Philip Papadopoulos, UCI • Tom DeFanti, UC San Diego Calit2/QI, • Frank Wuerthwein, UCSD Physics and SDSC Source: John Hess, CENIC Supercomputer Centers ESnet: Given Fast Networks, Need DMZs and Fast/Tuned DTNs
- 3. Terminating the Fiber Optics - Data Transfer Nodes (DTNs): Flash I/O Network Appliances (FIONAs) UCSD-Designed FIONAs Solved the Disk-to-Disk Data Transfer Problem at Near Full Speed on Best-Effort 10G, 40G and 100G Networks FIONAs Designed by UCSD’s Phil Papadopoulos, John Graham, Joe Keefe, and Tom DeFanti Two FIONA DTNs at UC Santa Cruz: 40G & 100G Up to 192 TB Rotating Storage Add Up to 8 Nvidia GPUs Per 2U FIONA To Add Machine Learning Capability
- 4. 2018/2019: Installing Community Shared FIONA CPU/GPU/Storage Systems on Campuses and Working With Campus CIOs on DMZs UC Merced Stanford UC Santa Barbara UC Riverside UC Santa Cruz UC Irvine Ann Kovalchick Panel 2
- 5. 2017-2020: CHASE-CI Grant Adds a Machine Learning Layer Built on Top of the Pacific Research Platform Caltech UCB UCI UCR UCSD UCSC Stanford MSU UCM SDSU NSF Grant for 256 High Speed “Cloud” GPUs For 32 ML Faculty & Their Students at 10 Campuses To Train AI Algorithms on Big Data
- 6. Original PRP CENIC/PW Link 2018-2021: Toward the National Research Platform (NRP) - Using CENIC & Internet2 to Connect Quilt Regional R&E Networks “Towards The NRP” 3-Year Grant Funded by NSF $2.5M October 2018 PI Smarr Co-PIs Altintas Papadopoulos Wuerthwein Rosing DeFanti NSF CENIC Link Original PRP CENIC/PW Link Sana Bellamine Panel 2 Jim Kyriannis Panel 2
- 7. 2018/2019: PRP Game Changer! Using Kubernetes to Orchestrate Containers Across the PRP User Applications Containers Clouds
- 8. PRP’s Nautilus Hypercluster Adopted Kubernetes to Orchestrate Software Containers and Rook, Which Runs Inside of Kubernetes, to Manage Distributed Storage https://rook.io/ “Kubernetes with Rook/Ceph Allows Us to Manage Petabytes of Distributed Storage and GPUs for Data Science, While We Measure and Monitor Network Use.” --John Graham, Calit2/QI UC San Diego
- 9. Rotating Storage 4000 TB PRP’s Nautilus is a Multi-Institution Hypercluster Connected by Optical Networks 180 FIONAs on 25 Partner Campuses Networked Together at 10-100Gbps
- 10. PRP Principles: Technology Evolution and Community Building • Innovation – Identify technologies nearing tipping points & evolve them into PRP stable services – Early access to new hardware is critical (FPGA/GPU/TPU/DPU…) – Aggressively adopt automation software systems • Adoption – Early engagement with opensource software projects ( Rook / Admiralty … ) – Share experience with the community (Workshops, Portal docs, Matrix chat, and weekly PRP call) • Scaling – Reduce complexity for initial onboarding of hardware, projects and users – Build more turnkey solutions supporting both research and instructional computing needs • Sustaining – Actively seek new collaborations – Grow the Nautilus user support service Social Network in the Matrix chat federation – Promote the Pot-Luck-Supercomputing paradigm John Graham Panel 2
- 11. PRP is Science-Driven: Connecting Multi-Campus Application Teams and Devices Earth Sciences UC San Diego UCBerkeley UC Merced Frank Wuerthwein Panel 1 Scott Sellars Panel 1 Alex Feltus Panel 1 Jeff Weekley Panel 1
- 12. Supporting and Expanding Emerging High-Priority PRP Applications – Climate Change-Induced Natural Hazards Prediction and Mitigation – Wildfires – Data Analysis from Large-Scale Scientific Instruments – CASPER – Community Biological Database Generation – Better Forcefields for COVID-19 Simulations – Understanding the Brain – NSF’s NeuroNex Ilkay Altintas Panel 1 Peak 1100 PRP CPU-Cores
- 13. Top 20 GPU Users Out of 400 Nautilus Namespace Applications: Together They Consumed Nearly 500 GPUs in 2020 Frank Wuerthwein, UCSD osggpus [IceCube] Mark Alber, UCR markalbergroup Nuno Vasconcelos, UCSD domain-adaptation Ravi Ramamoorthi, UCSD ucsd-ravigroup Hao Su, UCSD ucsd-haosulab Folding@Home folding Igor Sfiligoi, UCSD isfiligoi Xiaolong Wang, UCSD rl-multitask Xiaolong Wang, UCSD rl-multitask Xiaolong Wang, UCSD self-supervised-video Xiaolong Wang, UCSD hand-object-interaction Dinesh Bharadia, UCSD ecepxie Manmohan Chandraker, UCSD mc-lab Frank Wuerthwein, UCSD cms-ml Nuno Vasconcelos, UCSD svcl-oowl Vineet Bafna, UCSD ecdna Larry Smarr, UCSD jupyterlab Rose Yu, UCSD deep-forecast Nuno Vasconcelos, UCSD svcl-multimodal-learning Gary Cottrell, UCSD guru-research
- 14. Community Building Through Large-Scale Workshops Ana Hunsinger Futures
- 15. Increasing Participation Through PRP Science Engagement Workshops Source: Camille Crittenden, UC Berkeley UC Merced UC Davis UC Berkeley UC San Diego Chris Hoffman Futures
- 16. PRP Weekly Engineering Zoom Meeting Tom DeFanti Futures
- 17. Community Building Though Inclusion and Diversity: Workshops With Minority Serving Universities Richard Alo Panel 2