© 2001 Week (14 March 2001)Paradyn & Dyninst Demonstrations Paradyn & Dyninst Demos Barton P. Miller Computer.

Slides:



Advertisements
Similar presentations
Full-System Timing-First Simulation Carl J. Mauer Mark D. Hill and David A. Wood Computer Sciences Department University of Wisconsin—Madison.
Advertisements

Beowulf Supercomputer System Lee, Jung won CS843.
Last update: August 9, 2002 CodeTest Embedded Software Verification Tools By Advanced Microsystems Corporation.
Chapter 4 Threads, SMP, and Microkernels Patricia Roy Manatee Community College, Venice, FL ©2008, Prentice Hall Operating Systems: Internals and Design.
NUMA Tuning for Java Server Applications Mustafa M. Tikir.
Introduction to Operating Systems CS-2301 B-term Introduction to Operating Systems CS-2301, System Programming for Non-majors (Slides include materials.
Operating Systems Parallel Systems (Now basic OS knowledge)
CS 284a, 7 October 97Copyright (c) , John Thornley1 CS 284a Lecture Tuesday, 7 October 1997.
Operating Systems Parallel Systems and Threads (Soon to be basic OS knowledge)
OCT1 Principles From Chapter One of “Distributed Systems Concepts and Design”
AN INTRODUCTION TO LINUX OPERATING SYSTEM Zihui Han.
Microkernels, virtualization, exokernels Tutorial 1 – CSC469.
9/13/20151 Threads ICS 240: Operating Systems –William Albritton Information and Computer Sciences Department at Leeward Community College –Original slides.
Research Achievements Kenji Kaneda. Agenda Research background and goal Research background and goal Overview of my research achievements Overview of.
Tools and Utilities for parallel and serial codes in ENEA-GRID environment CRESCO Project: Salvatore Raia SubProject I.2 C.R. ENEA-Portici. 11/12/2007.
Chapter 4 Threads, SMP, and Microkernels Patricia Roy Manatee Community College, Venice, FL ©2008, Prentice Hall Operating Systems: Internals and Design.
Fall 2000M.B. Ibáñez Lecture 01 Introduction What is an Operating System? The Evolution of Operating Systems Course Outline.
A Virtual Machine Monitor for Utilizing Non-dedicated Clusters Kenji Kaneda Yoshihiro Oyama Akinori Yonezawa (University of Tokyo)
March 17, 2005 Roadmap of Upcoming Research, Features and Releases Bart Miller & Jeff Hollingsworth.
Chapter 2 Operating System Overview
O AK R IDGE N ATIONAL L ABORATORY U. S. D EPARTMENT OF E NERGY 1 On-line Automated Performance Diagnosis on Thousands of Processors Philip C. Roth Future.
Architectural Characterization of an IBM RS6000 S80 Server Running TPC-W Workloads Lei Yang & Shiliang Hu Computer Sciences Department, University of.
Architectural Characterization of an IBM RS6000 S80 Server Running TPC-W Workloads Lei Yang & Shiliang Hu Computer Sciences Department, University of.
March 12, 2001 Kperfmon-MP Multiprocessor Kernel Performance Profiling Alex Mirgorodskii Computer Sciences Department University of Wisconsin.
OPERATING SYSTEM SUPPORT DISTRIBUTED SYSTEMS CHAPTER 6 Lawrence Heyman July 8, 2002.
Computer Science Lecture 7, page 1 CS677: Distributed OS Multiprocessor Scheduling Will consider only shared memory multiprocessor Salient features: –One.
© 2004 Andrew R. BernatApril 14, 2004Dynamic Call-Path Profiling Incremental Call-Path Profiling Andrew Bernat Computer Sciences Department.
Debugging parallel programs. Breakpoint debugging Probably the most widely familiar method of debugging programs is breakpoint debugging. In this method,
BridgePoint Integration John Wolfe / Robert Day Accelerated Technology.
© 2002 Barton P. MillerMarch 4, 2001Tool Dæmon Protocol The Tool Dæmon Protocol: Using Monitoring Tools on Remote Applications Barton P. Miller
UNICOS. When it comes to solving real-world problems, leading-edge hardware is only part of the solution. A complete solution also requires a powerful.
© 2001 Barton P. MillerParadyn/Condor Week (12 March 2001, Madison/WI) The Paradyn Port Report Barton P. Miller Computer Sciences Department.
Preparatory Research on Performance Tools for HPC HCS Research Laboratory University of Florida November 21, 2003.
Lecture 5: Threads process as a unit of scheduling and a unit of resource allocation processes vs. threads what to program with threads why use threads.
Dynaprof Evaluation Report Adam Leko, Hans Sherburne UPC Group HCS Research Laboratory University of Florida Color encoding key: Blue: Information Red:
April 14, 2004 The Distributed Performance Consultant: Automated Performance Diagnosis on 1000s of Processors Philip C. Roth Computer.
PTools Annual Meeting, Knoxville, TN, September 2002 The Tool Daemon Protocol: Defining the Interface Between Tools and Process Management Systems.
Multi-core processors. 2 Processor development till 2004 Out-of-order Instruction scheduling Out-of-order Instruction scheduling.
Department of Computer Science and Software Engineering
Paradyn Project Paradyn / Dyninst Week Madison, Wisconsin April 12-14, 2010 Binary Rewriting with Dyninst Madhavi Krishnan and Dan McNulty.
Dynamic Instrumentation of Loops in Paradyn & Dyninst Eli Collins Computer Sciences Department University of Wisconsin-Madison Madison,
© 2001 B. P. Miller & M. Livny (12-14 March 2001)Paradyn/Condor Week Agenda Paradyn/Condor Week 2001 Barton P. Miller Miron Livny
Lecture 1: Network Operating Systems (NOS) An Introduction.
SIMULATION OF MULTIPROCESSOR SYSTEM AND NETWORK Manish Patel Nov 8 th 2004 Advisor: Dr. Chung-E-Wang Department of Computer Science California State University,
WASP Airborne Data Processor Laboratory for Imaging Algorithms and Systems Chester F. Carlson Center for Imaging Science Rochester Institute of Technology.
Chapter 1 Basic Concepts of Operating Systems Introduction Software A program is a sequence of instructions that enables the computer to carry.
Disco: Running Commodity Operating Systems on Scalable Multiprocessors Presented by: Pierre LaBorde, Jordan Deveroux, Imran Ali, Yazen Ghannam, Tzu-Wei.
Background Computer System Architectures Computer System Software.
Introduction Goal: connecting multiple computers to get higher performance – Multiprocessors – Scalability, availability, power efficiency Job-level (process-level)
PERFORMANCE OF THE OPENMP AND MPI IMPLEMENTATIONS ON ULTRASPARC SYSTEM Abstract Programmers and developers interested in utilizing parallel programming.
CT101: Computing Systems Introduction to Operating Systems.
Towards a High Performance Extensible Grid Architecture Klaus Krauter Muthucumaru Maheswaran {krauter,
Introduction to Operating Systems Concepts
Introduction to threads
Kernel Code Coverage Nilofer Motiwala Computer Sciences Department
Why to use the assembly and why we need this course at all?
Reliable Sockets: A Foundation for Mobile Communications
CompSci 725 Presentation by Siu Cho Jun, William.
Chapter 4: Threads.
Operating System Concepts
Chapter 4: Threads.
Department of Computer Science University of California, Santa Barbara
Chapter 4: Threads.
New Features in Dyninst 6.1 and 6.2
QNX Technology Overview
Efficient x86 Instrumentation:
Multithreaded Programming
Operating Systems (CS 340 D)
Department of Computer Science University of California, Santa Barbara
A Virtual Machine Monitor for Utilizing Non-dedicated Clusters
Presentation transcript:

© 2001 Week (14 March 2001)Paradyn & Dyninst Demonstrations Paradyn & Dyninst Demos Barton P. Miller Computer Sciences Department University of Wisconsin 1210 W. Dayton St. Madison, WI USA

Paradyn/Condor Week Demonstrations (14 March 2001)© 2001 Deep Start: A Hybrid Automated Performance Diagnosis Strategy We have developed Deep Start, a hybrid strategy for performing automated searches for application performance problems. The Deep Start strategy augments an automated search using dynamic instrumentation with stack sampling information. Using the stack sampling information, the Deep Start search strategy can focus its attention “closer” to the actual application bottlenecks, making the search more efficient as code size scales. In the demo, we will show how the Deep Start search strategy can improve the time required to find application bottlenecks in an MPI-based simulation. Philip Roth, University of Wisconsin - Madison

Paradyn/Condor Week Demonstrations (14 March 2001)© 2001 Heterogeneous MPI The Paradyn Parallel Performance Tool supports the monitoring and performance diagnosis for MPI applications built using the MPICH implementation. In this demo, we show Paradyn monitoring an MPI application running on x86 Linux, x86 Solaris, and SPARC Solaris. Paradyn Project, University of Wisconsin - Madison

Paradyn/Condor Week Demonstrations (14 March 2001)© 2001 Thread-aware Paradyn We have developed techniques to dynamically instrument threaded applications, and extended Paradyn to profile threaded programs. As a result we can provide detailed performance information for a threaded program without requiring special compiling and linking of the program. In this demo, we will show how the thread-aware Paradyn can provide detailed performance information for a multithreaded application running under SPARC Solaris. Brandon Schendel, University of Wisconsin - Madison

Paradyn/Condor Week Demonstrations (14 March 2001)© 2001 Paradyn: MIPS/IRIX Paradyn now supports hardware process CPU timers for applications running on MIPS IRIX systems. We are demonstrating Paradyn with an MPI application (built with IRIX native MPI) running on an SGI Origin Will Benton and Jeff Shergalis, University of Wisconsin - Madison {willb,

Paradyn/Condor Week Demonstrations (14 March 2001)© 2001 Paradyn: Power/AIX Paradyn on AIX 4.3 now supports shared-memory sampling and shared objects. We are demonstrating Paradyn with an MPI application, running under POE on the department’s new SP system. Drew Bernat, University of Wisconsin-Madison

Paradyn/Condor Week Demonstrations (14 March 2001)© 2001 Multiprocessor Kernel Performance Monitor As SMP systems become increasingly common in database, web, and file server contexts, kernel developers need performance tools that work with today’s (and tomorrow’s) multiprocessor-enabled kernels. Because threads may migrate from processor to processor throughout their lifetimes, the efficient collection of data from kernels running on multiprocessor systems is much more difficult than on uniprocessor systems. In this demo, we present a performance monitor tool that efficiently monitors the Solaris 7 kernel running on a sixteen-processor SPARC system. Alex Mirgorodskii, University of Wisconsin-Madison

Paradyn/Condor Week Demonstrations (14 March 2001)© 2001 Kernel Code Coverage Code coverage tools are effective for showing software developers which sections of their code are not executed. Applying this technology to a commodity operating system kernel presents the significant challenge of how to perform mass instrumentation. In this demo, we present a code coverage tool built using KernInst technology monitoring an unmodified SPARC Solaris 7 kernel. Nilofer Motiwala, University of Wisconsin-Madison

Paradyn/Condor Week Demonstrations (14 March 2001)© 2001 Reliable Sockets Reliable sockets (rocks) protect applications from network connection failures, include those due to Link failure IP address change Extended periods of disconnection In this demo, we show the use of rocks with ordinary shell and GUI- based applications. Victor Zandy, University of Wisconsin - Madison

Paradyn/Condor Week Demonstrations (14 March 2001)© 2001 DyninstAPI on Palm-sized PCs Running Windows CE As Palm-sized PCs become more widely used, they present an attractive target for tools using dynamic instrumentation technology. However, the Palm-sized PC target environment raises several significant challenges for DyninstAPI tool developers. In this demo, we present a tool that is a DyninstAPI mutator for processes running on a Windows CE device. Chadd Williams, University of Maryland