Parallelization of CC Workshop Benchmark Suggestion Sudhakar Pamidighantam NCSA.

Slides:



Advertisements
Similar presentations
Communication-Avoiding Algorithms Jim Demmel EECS & Math Departments UC Berkeley.
Advertisements

Write an equation given the slope and a point
CS2100 Computer Organisation Performance (AY2014/2015) Semester 2.
Autonomic Systems Justin Moles, Winter 2006 Enabling autonomic behavior in systems software with hot swapping Paper by: J. Appavoo, et al. Presentation.
EXAMPLE 5 Write and solve an equation
SAN DIEGO SUPERCOMPUTER CENTER Advanced User Support Project Outline October 9th 2008 Ross C. Walker.
Chapter 4 M. Keshtgary Spring 91 Type of Workloads.
Lincoln University Canterbury New Zealand Evaluating the Parallel Performance of a Heterogeneous System Elizabeth Post Hendrik Goosen formerly of Department.
A Parallel Computational Model for Heterogeneous Clusters Jose Luis Bosque, Luis Pastor, IEEE TRASACTION ON PARALLEL AND DISTRIBUTED SYSTEM, VOL. 17, NO.
CSCE 212 Chapter 4: Assessing and Understanding Performance Instructor: Jason D. Bakos.
EXAMPLE 1 Apply the distributive property
EXAMPLE 4 Writing and Evaluating an Expression Heart Rate a. Use n to write an expression for heart rate in beats per minute. b. After exercising, you.
Measuring zSeries System Performance Dr. Chu J. Jong School of Information Technology Illinois State University 06/11/2012 Sponsored in part by Deer &
Copyright © 2013 by The McGraw-Hill Companies, Inc. All rights reserved.McGraw-Hill/Irwin.
SEC(R) 2008 Intel® Concurrent Collections for C++ - a model for parallel programming Nikolay Kurtov Software and Services.
SOS7, Durango CO, 4-Mar-2003 Scaling to New Heights Retrospective IEEE/ACM SC2002 Conference Baltimore, MD Distilled [Trimmed & Distilled for SOS7 by M.
Creating a Parallel Program to Compute Statistical Information Victoria Sensano Maui Scientific Research Center Research Supervisor: Douglas Hope.
Scaling to New Heights Retrospective IEEE/ACM SC2002 Conference Baltimore, MD.
So, Jung-ki Distributed Computing System LAB School of Computer Science and Engineering Seoul National University Implementation of Package Management.
Unifying Primary Cache, Scratch, and Register File Memories in a Throughput Processor Mark Gebhart 1,2 Stephen W. Keckler 1,2 Brucek Khailany 2 Ronny Krashinsky.
Software Performance Analysis Using CodeAnalyst for Windows Sherry Hurwitz SW Applications Manager SRD Advanced Micro Devices Lei.
Physical Database Design & Performance. Optimizing for Query Performance For DBs with high retrieval traffic as compared to maintenance traffic, optimizing.
Ideas to Improve SharePoint Usage 4. What are these 4 Ideas? 1. 7 Steps to check SharePoint Health 2. Avoid common Deployment Mistakes 3. Analyze SharePoint.
Service Computation 2010November 21-26, Lisbon.
Taking the Complexity out of Cluster Computing Vendor Update HPC User Forum Arend Dittmer Director Product Management HPC April,
Thread-Level Speculation Karan Singh CS
EXAMPLE 3 Solving an Equation with Mixed Numbers Biology
Example 3 Using Mixed Numbers in Real Life A corn snake that is inches long grows to a length of inches. How much does it grow? BIOLOGY
By Garrett Kelly. 3 types or reasons for distributed applications Data Data used by the application is distributed Computation Computation is distributed.
1 " Teaching Parallel Design Patterns to Undergraduates in Computer Science” Panel member SIGCSE The 45 th ACM Technical Symposium on Computer Science.
Cracow Grid Workshop October 2009 Dipl.-Ing. (M.Sc.) Marcus Hilbrich Center for Information Services and High Performance.
Scaling Area Under a Curve. Why do parallelism? Speedup – solve a problem faster. Accuracy – solve a problem better. Scaling – solve a bigger problem.
Georgia Institute of Technology, Microelectronics Research Center Prediction of Interconnect Fan-out Distribution Using Rent’s Rule Payman Zarkesh-Ha,
Copyright © 2009 Pearson Education, Inc. Publishing as Pearson Addison-Wesley Principles of Parallel Programming First Edition by Calvin Lin Lawrence Snyder.
Computer and Computational Sciences Division Los Alamos National Laboratory On the Feasibility of Incremental Checkpointing for Scientific Computing Jose.
High Performance Computing on an IBM Cell Processor Bioinformatics Team Members Kyle Byerly Shannon McCormick Matt Rohlf Bryan Venteicher Advisor Dr. Zhao.
Lecture 2 Conditional Statement. chcslonline.org Conditional Statements in PHP Conditional Statements are used for decision making. Different actions.
Department of Computer Science MapReduce for the Cell B. E. Architecture Marc de Kruijf University of Wisconsin−Madison Advised by Professor Sankaralingam.
Workshop on Parallelization of Coupled-Cluster Methods Panel 1: Parallel efficiency An incomplete list of thoughts Bert de Jong High Performance Software.
Parallel Portability and Heterogeneous programming Stefan Möhl, Co-founder, CSO, Mitrionics.
Parallelizing Spacetime Discontinuous Galerkin Methods Jonathan Booth University of Illinois at Urbana/Champaign In conjunction with: L. Kale, R. Haber,
6/29/1999PDPTA'991 Performance Prediction for Large Scale Parallel Systems Yuhong Wen and Geoffrey C. Fox Northeast Parallel Architecture Center (NPAC)
EXAMPLE 1 Apply the distributive property
COMP381 by M. Hamdi 1 Clusters: Networks of WS/PC.
Lx: A Technology Platform for Customizable VLIW Embedded Processing.
1 Parallel Applications Computer Architecture Ning Hu, Stefan Niculescu & Vahe Poladian November 22, 2002.
GridChem Developers Conference Focus For Final Year Sudhakar Pamidighantam NCSA 25 August 2006.
GridChem Sciene Gateway and Challenges in Distributed Services Sudhakar Pamidighantam NCSA, University of Illinois at Urbaba- Champaign
Scaling Conway’s Game of Life. Why do parallelism? Speedup – solve a problem faster. Accuracy – solve a problem better. Scaling – solve a bigger problem.
CMSC 611: Advanced Computer Architecture Performance & Benchmarks Some material adapted from Mohamed Younis, UMBC CMSC 611 Spr 2003 course slides Some.
2 nd Austrian HPC Workshop Heuristiclab Hive Goals Realization Deployment Page1.
Sec Math II 1.3.
FTC-Charm++: An In-Memory Checkpoint-Based Fault Tolerant Runtime for Charm++ and MPI Gengbin Zheng Lixia Shi Laxmikant V. Kale Parallel Programming Lab.
COMP7500 Advanced Operating Systems I/O-Aware Load Balancing Techniques Dr. Xiao Qin Auburn University
1 Potential for Parallel Computation Chapter 2 – Part 2 Jordan & Alaghband.
CSE 340 Computer Architecture Summer 2016 Understanding Performance.
4- Performance Analysis of Parallel Programs
. . .
Excitable Cell Simulation In Star-P
Auburn University COMP7500 Advanced Operating Systems I/O-Aware Load Balancing Techniques (2) Dr. Xiao Qin Auburn University.
CSCE 212 Chapter 4: Assessing and Understanding Performance
BACK SOLUTION:
EXAMPLE 2 Identify parallel lines
Counting counts Quantifiable methods and data easy to learn
مديريت موثر جلسات Running a Meeting that Works
Hybrid Programming with OpenMP and MPI
Distributed Computing:
Case Studies with Projections
Systems of Linear Equations: An Introduction
Simplify by combining like terms
Presentation transcript:

Parallelization of CC Workshop Benchmark Suggestion Sudhakar Pamidighantam NCSA

General Benchmark needs Benchmarks standardization is important but comparison with/between codes could be problematic if license statements prohibit such activity The information should provide users a guide lines to their own parallel runs The benchmarks between heterogeneous systems may not be comparable except for total time to solution Benchmarking is used to evaluate systems for price/performance and should be a continuous process NSF has a set that could be a start if we want one

Heterogeneous systems GPU/CPU/FPGA count and their usage Cache amounts and bandwidths

Goal for Benchmarks Automatic processor count/type selection Problem specificity On the fly benchmarking for specific distribution cpu/io

Systems NSF Set Natural systems easy to systematically grow like benzene--- hexacene…./Polymer/Argon clusters to define some constant work/data per processing unit Method dependent