Deploying a Top-100 Supercomputer for Large Parallel Workloads: the Niagara Supercomputer

Abstract

Niagara is currently the fastest supercomputer accessible to academics in Canada. It was deployed at the beginning of 2018 and has been serving the research community ever since. This homogeneous 60,000-core cluster, owned by the University of Toronto and operated by SciNet, was intended to enable large parallel jobs and has a measured performance of 3.02 petaflops, debuting at #53 in the June 2018 TOP500 list. It was designed to optimize throughput of a range of scientific codes running at scale, energy efficiency, and network and storage performance and capacity. It replaced two systems that SciNet operated for over 8 years, the Tightly Coupled System (TCS) and the General Purpose Cluster (GPC). In this paper we describe the transition process from these two systems, the procurement and deployment processes, as well as the unique features that make Niagara a one-of-a-kind machine in Canada.

PEARC'19: "Practice and Experience in Advanced Research Computing", July 28-August 1, 2019, Chicago, IL, USA

References in corpus (1)

• Simulating 3-D Stellar Hydrodynamics using PPM and PPB Multifluid Gas Dynamics on CPU and CPU+GPU Nodes

Cited by in corpus (31)

• The Websky Extragalactic CMB Simulations
• Quantum Computational Advantage via High-Dimensional Gaussian Boson Sampling
• Fault-tolerant quantum computation with static linear optics
• Exploring the Sgr-Milky-Way-disc interaction using high resolution N-body simulations
• Jet: Fast quantum circuit simulations with parallel task-based tensor-network contraction
• Natural Evolutionary Strategies for Variational Quantum Computation
• Unsupervised Hadronic SUEP at the LHC
• On the road to percent accuracy III: non-linear reaction of the matter power spectrum to massive neutrinos
• LBfoam: An open-source software package for the simulation of foaming using the Lattice Boltzmann Method
• Determination of Trace Organic Contaminant Concentration via Machine Classification of Surface-Enhanced Raman Spectra
• Noise robustness and experimental demonstration of a quantum generative adversarial network for continuous distributions
• Localizing FRBs through VLBI with the Algonquin Radio Observatory 10-m Telescope
• CMB lensing reconstruction biases from masking extragalactic sources
• General Relativistic Hydrodynamics on a Moving-mesh I: Static Spacetimes
• An In-Depth Analysis of the Slingshot Interconnect
• A unifying algebraic framework for discontinuous Galerkin and flux reconstruction methods based on the summation-by-parts property
• General relativistic precession and the long-term stability of the solar system
• RACS: Rapid Analysis of ChIP-Seq data for contig based genomes
• Information flow in parameterized quantum circuits
• A Parallel Boundary Element Method for the Electromagnetic Analysis of Large Structures With Lossy Conductors
• Confinement on ${\mathbb{R}}^3\times {\mathbb{S}}^1$ and Double-String Collapse
• Constraining Ultralight Axions with Galaxy Surveys
• The XFaster Power Spectrum and Likelihood Estimator for the Analysis of Cosmic Microwave Background Maps
• Using Giant Pulses to Measure the Impulse Response of the Interstellar Medium
• Statistical exploration of halo anisotropic clustering and intrinsic alignments with the mass-Peak Patch algorithm
• The Hydrogen Intensity and Real-time Analysis eXperiment: 256-Element Array Status and Overview
• Stochastic RWA and Lightpath Rerouting in WDM Networks
• Gemini: Dynamic Bias Correction for Autonomous Experimentation and Molecular Simulation
• Collapsar Disk Outflows III: Detectable Neutrino and Gravitational Wave Signatures
• Parallel Ray Tracing of Black Hole Images Using the Schwarzschild Metric
• Resolving the fastest ejecta from binary Neutron Star mergers: implications for electromagnetic counterparts