High Performance Computing University of Southern California

Slides:



Advertisements
Similar presentations
© 2012 Open Grid Forum Simplifying Inter-Clouds October 10, 2012 Hyatt Regency Hotel Chicago, Illinois, USA.
Advertisements

The ADAMANT Project: Linking Scientific Workflows and Networks “Adaptive Data-Aware Multi-Domain Application Network Topologies” Ilia Baldine, Charles.
 Amazon Web Services announced the launch of Cluster Compute Instances for Amazon EC2.  Which aims to provide high-bandwidth, low- latency instances.
Programming in the Many Software Engineering Paradigm for the 21 st Century Nenad Medvidovic Center for Software Engineering Computer Science Department.
Deconstructing Clusters for High End Biometric Applications NSF CCF June Douglas Thain and Patrick Flynn University of Notre Dame 5 August.
Software Engineering COMP 201
SYSTEMS SUPPORT FOR GRAPHICAL LEARNING Ken Birman 1 CS6410 Fall /18/2014.
Iterative computation is a kernel function to many data mining and data analysis algorithms. Missing in current MapReduce frameworks is collective communication,
Teaching Fellow Admissions Tutor for Computer Science Director of Undergraduate Studies.
SCD FIFE Workshop - GlideinWMS Overview GlideinWMS Overview FIFE Workshop (June 04, 2013) - Parag Mhashilkar Why GlideinWMS? GlideinWMS Architecture Summary.
Abstract Load balancing in the cloud computing environment has an important impact on the performance. Good load balancing makes cloud computing more.
Programming Models & Runtime Systems Breakout Report MICS PI Meeting, June 27, 2002.
COMS E Cloud Computing and Data Center Networking Sambit Sahu
The Future of the iPlant Cyberinfrastructure: Coming Attractions.
Alternative ProcessorsHPC User Forum Panel1 HPC User Forum Alternative Processor Panel Results 2008.
Software Defined Networks for Dynamic Datacenter and Cloud Environments.
Distributed Information Systems. Motivation ● To understand the problems that Web services try to solve it is helpful to understand how distributed information.
AFRD modeling and simulation meeting – 09/09/2013 Introduction - J.-L. Vay Snowmass CSS 2013 – Computing Frontier: accelerator science.
Design Patterns -- Omkar. Introduction  When do we use design patterns  Uses of design patterns  Classification of design patterns  Creational design.
Monitoring Business Processes with Queries VLDB2007 CatrielBeeri, AnatEyal, Tova Milo, AlonPilberg Summarized by Gong GI Hyun, IDS Lab., Seoul.
Operating Systems: Wrap-Up Questions answered in this lecture: What is an Operating System? Why are operating systems so interesting? What techniques can.
Upcoming GENI Architecture Topics: The Future of Experiment Management with Gush Jeannie Albrecht David Irwin.
DiRAC-3 – The future Jeremy Yates, STFC DiRAC HPC Facility.
Experiences Running Seismic Hazard Workflows Scott Callaghan Southern California Earthquake Center University of Southern California SC13 Workflow BoF.
CISC Machine Learning for Solving Systems Problems Presented by: Eunjung Park Dept of Computer & Information Sciences University of Delaware Solutions.
Windows Azure poDRw_Xi3Aw.
Presented by The Harness Workbench: Unified and Adaptive Access to Diverse HPC Platforms Christian Engelmann Computer Science Research Group Computer Science.
SCEC: An NSF + USGS Research Center Focus on Forecasts Motivation.
Hongbin Li 11/13/2014 A Debugger of Parallel Mutli- Agent Spatial Simulation.
IFETCE/ME/CSE/B.V.R.Raju/Iyear/Isem/CP7103/MCA/Unit-4/PPt/Ver1.0
 Programming methodology: ◦ is a process of developing programs that involves strategically dividing important tasks into functions to be utilized by.
Prof. Jong-Moon Chung’s Lecture Notes at Yonsei University
Seminar Announcement December 24, Saturday, 15:00-17:00, Room: A302, WNLO Title: Quality-of-Experience (QoE) and Power Efficiency Tradeoff for Fog Computing.
Accessing the VI-SEEM infrastructure
Jan Odegard, Ken Kennedy Institute for Information Technology, Rice
Introduction: Computer programming
SPIDAL Analytics Performance February 2017
Dynamo: A Runtime Codesign Environment
Joslynn Lee – Data Science Educator
A Cloudy Future of What? Jeff Hollingsworth.
Example: Rapid Atmospheric Modeling System, ColoState U
Dagstuhl Seminar on Dark Silicon: From Embedded to HPC Feb 3, 2016
Operating System Structure
Abstract Major Cloud computing companies have started to integrate frameworks for parallel data processing in their product portfolio, making it easy for.
Introduction to XSEDE Resources HPC Workshop 08/21/2017
Systematic Manual Testing
Data Center Energy Efficiency: Scale-Up/Scale-Out Processor Design Background & Analysis By Nick.
Biology MDS and Clustering Results
Tonga Institute of Higher Education
Martin Swany Gregor von Laszewski Thomas Sterling Clint Whaley
Scalable Parallel Interoperable Data Analytics Library
Alternative Processor Panel Results 2008
Jigar.B.Katariya (08291A0531) E.Mahesh (08291A0542)
$1M a year for 5 years; 7 institutions Active:
Key Manager Domains February, 2019.
Why Threads Are A Bad Idea (for most purposes)
Mark McKelvin EE249 Embedded System Design December 03, 2002
Overview of Workflows: Why Use Them?
TensorFlow: A System for Large-Scale Machine Learning
Why Threads Are A Bad Idea (for most purposes)
Why Threads Are A Bad Idea (for most purposes)
Big Data, Simulations and HPC Convergence
DiRAC Technical Application
Welcome to (HT)Condor Week #19 (year 34 of our project)
Mobile Computing With Android ACST 4550
Welcome to HTCondor Week #17 (year 32 of our project)
Convergence of Big Data and Extreme Computing
Object Oriented Design
AWS Computing NTEG June 2019 New Technology Exploration Group.
Presentation transcript:

High Performance Computing University of Southern California Viktor K. Prasanna University of Southern California prasanna@usc.edu NSF CSR PI Meet June 2, 2017

High Performance Computing (1) What is HPC, what are interesting applications? New architectures are becoming complex, how does users program them? Optimize for them? Maintain them? Simplified models—resource oblivious programming? Code adaptation—complier? run time? Performance portability across architectures Debugging tools HPC on the Cloud? can we reduce overheads to make Clouds attractive for HPC workloads HPC at the Edge?

High Performance Computing (2) 4) Fundamental understanding of HPC systems? Too many abstractions? 5) Changing HPC workloads data science new architectures to support these 6) Access to HPC resources overheads in obtaining access access to large scale machines 7) NSF, DoE: where does the community submit?