We'll explain why analyzing these vast networks with possibly billions of entries requires the computing power of GPUs. It has proven ability to rapidly run very large, multi-million cell, full-physics models using massive parallelism.

We hig... Data sets such as graphs are growing so rapidly that performing meaningful data analytics in reasonable time is beyond the ability of common software and hardware for many applications. Joe joined NVIDIA in 2013 to work in the CUDA Libraries team on sparse linear algebra methods and especially the AmgX library of GPU-accelerated sparse iterative solvers. His fascination with CFD and fluid mechanics led to two Mechanical Engineering degrees (Rice University and Stanford University) before he decided it was really all about the math. Joe Eaton holds a Ph.D. in Computational and Applied Mathematics from the University of Texas at Austin's TICAM program. Joe EATON, Technical lead of NVIDIA, CA (Nvidia) | Read 14 publications | Contact Joe EATON Currently he is the architect and technical lead on nvGRAPH, which applies sparse linear algebra and machine learning techniques for graph theory problems. Joe Eaton NVIDIA ECHELON is one of the most disruptive technologies I’ve seen in my career doing simulation. Previous work have shown that GCNs are vulnerable to the perturbation on adjacency and feature matrices of existing nodes.

Understand dynamic and mechanical properties of Asiatic composite bows, improve these designs by modifying geometry and materials used. Substituting the basis representaton into equation (2), we obtain. Doctor of Philosophy Computational and Applied Mathematics. In the extreme case in which vertical permeability goes to zero, the Joshi model predicts a 0.0 stb/day-psi PI, which is wrong. All rights reserved. Joe Eaton is the technical lead for Accelerated Graph and Data Analytics at NVIDIA. A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. The subspace size has an important impact on the performances, however, it is se... Heterogeneous supercomputers combining “general-purpose graphic processor”, “many integrated cores” and general-purpose CPUs promise to be the future major architecture, because they deliver excellent performance with limited power consumption and space occupancy. In this paper,... Our approach for parallel multiphysics and multiscale simulation uses two levels of domain decomposition: physical and computational. Joe Eaton is the technical lead for Accelerated Graph and Data Analytics at NVIDIA.

NVIDIA websites use cookies to deliver and improve the website experience. Graph convolutional networks (GCNs) have been widely used for classifying graph nodes in the semi-supervised setting. Join ResearchGate to find the people and research you need to help your work. JOE Eaton U.S. Benefits. The implicitly restarted Arnoldi method (IRAM) is a method to compute a set of eigenpairs of large sparse general matrices based on Krylov subspace techniques. An isothermal, implicit, mixed finite element black oil reservoir simulator from the University of Texas is coupled to an explicit, quasistatic, nonlinear finite element solid mechanics code from Sandia National Laboratories.

Only verified researchers can join ResearchGate and send messages to other members. © 2008-2020 ResearchGate GmbH. AmgX: A Library for GPU Accelerated Algebraic Multigrid and Preconditioned Iterative Methods, Predicting Productivity Index of Horizontal Wells, A Parallel Multiblock/Multidomain Approach for Reservoir Simulation, Staggered In Time Coupling of Reservoir Flow Simulation and Geomechanical Deformation: Step 1 — One-Way Coupling, Numerical Solution of Partial Differential Equations using a Wavelet Basis, Staggered In Time Coupling of Reservoir Flow Simulation and Geomechanical Deformation: Step 1 - One-Way Coupling, Transport of multispecies contaminants with biological and chemical kinetics in porous media, Finite element models for reservoir simulation, Institut des Sciences et Techniques des Yvelines (ISTY), Laboratoire d'Informatique Fondamentale de Lille (LIFL), Technology, Computer Science and Applied Mathematics Division. See our cookie policy for further details on how we use cookies and how to change your cookie settings. Previously, he was chief scientist at Object Reservoir, producing finite-element based 3phase reservoir simulations with automated adaptive meshing technology. Joe Eaton (NVIDIA) Graphs are a ubiquitous part of technology we use daily in systems like GPS graphs help find the shortest path between two points and in social networks, which use them to help users find friends. He was also vice president of engineering at Algebraix Data, leading a development team to produce a high-performance SPARQL query engine.

Sorry, you need to be a researcher to join ResearchGate. He joined NVIDIA in 2013 to lead the AmgX product team. He joined NVIDIA in 2013 to lead the AmgX product team. In this paper we present a solut... We present a parallel, hybrid solver to compute a set of eigenpairs of large, sparse, non-symmetric matrices. In this context, performance and efficiency are primary concerns. In realistic cases the matrices are often so large that they require large scale distributed parallel computing to obtain the solution of interest in a reasonable time. AmgX Overview Two forms of AMG Classical AMG, as in HYPRE, strong convergence, scalar Un-smoothed Aggregation AMG, lower setup times, handles block systems Krylov methods We develop an efficient parallel algorithm for computing Jaccard edge and PageRank vertex weights. The popular Joshi model slightly overestimated the flow resistance of a horizontal well. Will GPGPUs be Finally a Credible Solution for Industrial Reservoir Simulators?

Both codes are 3d and parallel.

Joe is a competent leader, a treasure trove of technical facts, and an innovator. Previously, he was chief scientist at Object Reservoir, producing finite-element based 3phase reservoir simulations with automated adaptive meshing technology. University students and faculty, institute members, and independent researchers, Technology or product developers, R&D specialists, and government or NGO employees in scientific roles, Health care professionals, including clinical researchers, Journalists, citizen scientists, or anyone interested in reading and discovering research. Joe's Ph.D. work was on AMG applied to reservoir simulation problems, mixed with high performance chemistry simulation and parallel computing. In this paper we propose to generalize Jaccard and related measures, often used as similarity coefficients between two sets. Copyright © 2020 NVIDIA Corporation, Explore our regional blogs and other social networks, ARCHITECTURE, ENGINEERING AND CONSTRUCTION, How GPUs Are Transforming the Oil and Gas Industry. As a manager, he seeks to understand his team and facilitate them, eliminating road blocks, identifying … The spectral analysis of real networks reflects such problematic. We define Jaccard, Dice-Sorensen and Tversky edge weights on a graph and generalize them to account for vertex weights. Due to our privacy policy, only current members can send messages to people on ResearchGate. Joe Eaton talks about the state of AI and the cloud at the Nimbix Developer Summit 2017. Dr. Joe Eaton, NVIDIA .

Institute for Computational Engineering and Sciences, Research on subsurface flows, including chemically reacting and multi-phase flows. First, the physical domain is decomposed into subdomains or blocks according to the geometry, geology, and physics/chemistry/biology. Post by Joe Eaton Linear Solvers are Necessary CFD Energy Physics Nuclear Safety . The solution of large sparse linear systems arises in many applications, such as computational fluid dynamics and oil reservoir simulation. As a result of this, the Joshi model underpredicts the productivity index (PI) of a horizontal well by a few percent.

... Eaton’s 2019 revenues were $21.4 billion, and we sell products to customers in more than 175 countries. Emphasis on efficient solvers for the linear systems that arise in these models, National Institute for Research in Computer Science and Control, Retired with 30+ years from Sandia National Labs, Oden Institute for Computational Engineering and Sciences, Attack Graph Convolutional Networks by Adding Fake Nodes, Parallel jaccard and related graph clustering techniques, Accelerated Hybrid Approach for Spectral Problems Arising in Graph Analytics, Leveraging accelerators in the multiple implicitly restarted Arnoldi method with nested subspaces, Programming Perspectives for Pre-exascale Systems. See our, Parallel Direct Solvers with cuSOLVER: Batched QR, AmgX V1.0: Enabling Reservoir Simulation with Classical AMG, AmgX: Multi-Grid Accelerated Linear Solvers for Industrial Applications. Joe has experience in multi-phase and multi-physics simulators on large parallel computers, including convection-diffusion-reaction systems with millions of cells and tens of millions of chemistry unknowns. Each subdomain represents a single physical system, on a reasonable range of scale... this paper. Joe lives in Austin, Texas, and holds a Ph.D. in computational and applied mathematics from UT Austin, a master's in mechanical engineering from Stanford, and a bachelor's in mechanical engineering from Rice University. Joe's Ph.D. work was on AMG applied to reservoir simulation problems, mixed with high performance chemistry simulation and parallel computing. NVIDIA websites use cookies to deliver and improve the website experience.