Three Questions To Ask About Clusters

Slides:



Advertisements
Similar presentations
Three Questions You Should Ask William Gropp Mathematics and Computer Science.
Advertisements

What Is An Operating System?. Before Operating Systems u What do you do with just computer hardware? –If someone gives you a computer with no software.
Introductions to Parallel Programming Using OpenMP
CS 732: Advance Machine Learning Usman Roshan Department of Computer Science NJIT.
Threads CS 416: Operating Systems Design, Spring 2001 Department of Computer Science Rutgers University
Contemporary Languages in Parallel Computing Raymond Hummel.
INTRODUCTION TO CLOUD COMPUTING Cs 595 Lecture 5 2/11/2015.
1 CS503: Operating Systems Part 1: OS Interface Dongyan Xu Department of Computer Science Purdue University.
Virtualization Concept. Virtualization  Real: it exists, you can see it.  Transparent: it exists, you cannot see it  Virtual: it does not exist, you.
Parallel and Distributed Computing Overview and Syllabus Professor Johnnie Baker Guest Lecturer: Robert Walker.
CS 3305 Course Overview. Introduction r Instructor: Dr Hanan Lutfiyya r Office: MC 355 r hanan at csd dot uwo ca r Office Hours: m Drop-by m Appointment.
GCSE Computing Baxter College. Computing in Schools What is computing? Course Content Benefits Difference between ICT & Computing? TO DISCUSS.
9/13/20151 Threads ICS 240: Operating Systems –William Albritton Information and Computer Sciences Department at Leeward Community College –Original slides.
Additional Materials1 Other Languages C and C++: Languages used by Systems Programmers. Heavy use by Computer Scientists. COBOL: An ancient Business oriented.
1 Lecture 20: Parallel and Distributed Systems n Classification of parallel/distributed architectures n SMPs n Distributed systems n Clusters.
View-Oriented Parallel Programming for multi-core systems Dr Zhiyi Huang World 45 Univ of Otago.
U.S. Department of Energy’s Office of Science High Performance Computing Challenges and Opportunities Dr. Daniel Hitchcock
Operating Systems Lecture 2 Processes and Threads Adapted from Operating Systems Lecture Notes, Copyright 1997 Martin C. Rinard. Zhiqing Liu School of.
Chapter 2 (PART 1) Light-Weight Process (Threads) Department of Computer Science Southern Illinois University Edwardsville Summer, 2004 Dr. Hiroshi Fujinoki.
Tami Meredith, Ph.D. CSCI  Devices need CPU access  E.g., NIC has a full buffer it needs to empty  These device needs are often asynchronous.
Copyright © 2007 Pearson Education, Inc. Publishing as Pearson Addison-Wesley Slide 1- 1 Overview 1.1 Computer Systems 1.2 Programming and Problem Solving.
Topic 1Topic 2Topic 3Topic 4Topic
1 Programming Languages B.J. Maclennan 4. Syntax and Elagance: Algol-60.
Department of Computer Science Southern Illinois University Edwardsville Spring, 2010 Dr. Hiroshi Fujinoki IPC1.PPT/001 Inter-Process.
C o n f i d e n t i a l 1 Course: BCA Semester: III Subject Code : BC 0042 Subject Name: Operating Systems Unit number : 1 Unit Title: Overview of Operating.
CS 326: Functional Programming 1. 2 Erlang – A survey of the language & applications Paper by: Joe Armstrong, Computer Science Laboratory, Ericsson Telecom.
University of Maryland Towards Automated Tuning of Parallel Programs Jeffrey K. Hollingsworth Department of Computer Science University.
HW & Systems: Operating Systems IS 101Y/CMSC 101 Computational Thinking and Design Tuesday, October 22, 2013 Marie desJardins University of Maryland, Baltimore.
Parallel and Distributed Computing Overview and Syllabus Professor Johnnie Baker Guest Lecturer: Robert Walker.
What is Science?. The Goal of Science to investigate and understand the natural world To explain events in the natural world To use those explanations.
CS614: Advanced Course in Computer Systems (Spring’04) Instructor: Ken Birman TA: non assigned (yet)
CS 732: Advance Machine Learning
Ways to act like a computer scientist. Be curious Look at different menus Run code to see what happens Check the effects of different settings Answer.
CSCI N201 Programming Concepts and Database 2 - STAIR Lingma Acheson Department of Computer and Information Science, IUPUI.
Gauss Students’ Views on Multicore Processors Group members: Yu Yang (presenter), Xiaofang Chen, Subodh Sharma, Sarvani Vakkalanka, Anh Vo, Michael DeLisi,
Hardware Trends CSE451 Andrew Whitaker. Motivation Hardware moves quickly OS code tends to stick around for a while “System building” extends way beyond.
GPU Computing for GIS James Mower Department of Geography and Planning University at Albany.
Lecture 4 Page 1 CS 111 Online Modularity and Memory Clearly, programs must have access to memory We need abstractions that give them the required access.
1 University of Maryland Using Information About Cache Evictions to Measure the Interactions of Application Data Structures Bryan R. Buck Jeffrey K. Hollingsworth.
Parallel OpenFOAM CFD Performance Studies Student: Adi Farshteindiker Advisors: Dr. Guy Tel-Zur,Prof. Shlomi Dolev The Department of Computer Science Faculty.
Conclusions on CS3014 David Gregg Department of Computer Science
Why don’t programmers have to program in machine code?
OPERATING SYSTEMS CS 3502 Fall 2017
Chapter 13 The Function of Operating Systems
Resource Management IB Computer Science.
APIs and Function Parameters
Introduction.
Lesson Objectives Aims Key Words
OPERATING SYSTEMS CS3502 Fall 2017
Learning About Operating Systems
MOBILE DEVICE OPERATING SYSTEM
Modularity and Memory Clearly, programs must have access to memory
Intro to Multiprocessing
Type Topic in here! Created by Educational Technology Network
Computer Systems Summary
HW & Systems: Operating Systems IS 101Y/CMSC 101 Computational Thinking and Design Tuesday, October 22, 2013 Carolyn Seaman University of Maryland, Baltimore.
I can ask and answer text dependent questions
What is an Operating System?
Operating Systems (CS 340 D)
S.T.A.I.R CSCI N207 Data Analysis Using Spreadsheet Lingma Acheson
CS 286 Computer Architecture & Organization
Introduction to OS (concept, evolution, some keywords)
Organization- Unit 1 CDA
Type Topic in here! Created by Educational Technology Network
Introduction to OS (concept, evolution, some keywords)
Light-Weight Process (Threads)
Lecture Topics: 11/1 Hand back midterms
Why and How We Use the The Scientific Method.
Convergence of Big Data and Extreme Computing
CS Introduction to Operating Systems
Presentation transcript:

Three Questions To Ask About Clusters Jeffrey K. Hollingsworth hollings@cs.umd.edu Department of Computer Science University of Maryland, College Park, MD 20742

What goes bump in the night on the cluster? Can I get Performance within X% for two consecutive runs on the same data and cluster? What goes bump in the night on the cluster? OS Processes Power/Heat Scaling on the processor What changes when upgrades happen? Operating System Libraries Hardware Open Question: What value of X is right? 1%, 5%, 10%, 100%?

When Can I Stop Caring About the Network Topology (or Technology)? Range Of Answers: Today -> Never Many Computational Scientists Don’t Care Now Many CS Researchers Still Publish on this topic No Fair making the MPI stack slow to make this possible

How Much Time will I Save with your new Language and at what cost? Programmer Effort Often Today: Trust me it will be easier: Threads, HPF, Transactional Memory Need: Rigorous Study of benefits of features Performance How much performance to give up for easier programming?