The El-Rewini/Ali Scheduling of In-Forest Task Graph on Two Processors with Communication Project Presentation By David Y. Feinstein SMU - CSE 8388 Spring.

Slides:

Advertisements

Similar presentations

Quick Start Guide: Starting a PT eMeeting and Managing the User Directory This presentation assumes you are familiar with the unique capabilities and benefits.

Advertisements

© 2004 Wayne Wolf Topics Task-level partitioning. Hardware/software partitioning.  Bus-based systems.

With TimeCard SharePoint events are tagged with information that converts them into time sheets. This way users can report time and expenses from their.

 Use the Left and Right arrow keys or the Page Up and Page Down keys to move between the pages. You can also click on the pages to move forward.  To.

Evaluating Heuristics for the Fixed-Predecessor Subproblem of Pm | prec, p j = 1 | C max.

Schedule and Effort. Planning Big Project: Waterfall-ish Style 1.Figure out what the project entails Requirements, architecture, design 2.Figure out dependencies.

With TimeCard appointments are tagged with information that converts them into time sheets. This way users can report time and expenses from their Outlook.

Author: David He, Astghik Babayan, Andrew Kusiak By: Carl Haehl Date: 11/18/09.

Lab/Sessional -CSE-374. SYSTEM DEVELOPMENT LIFE CYCLE.

Computer Network Project Computer Network Project Efficient handling of messages with multimedia attachments.

1 Session-I & II CSIT-121 Spring 2006 Session Targets Introducing the VC++.NET Solving problems on computer Programming in C++ Writing and Running Programs.

Chapter 11 - Monitoring Server Performance1 Ch. 11 – Monitoring Server Performance MIS 431 – created Spring 2006.

1 Spring 2007 CSCI 660 CSCI-660 Project Title Project team members’ names.

CSE 421 Algorithms Richard Anderson Lecture 6 Greedy Algorithms.

On the Task Assignment Problem : Two New Efficient Heuristic Algorithms.

Using the Unity 98 Cordless Response System Directions : Courtesy of IHC public relations.

Threshold Voltage Assignment to Supply Voltage Islands in Core- based System-on-a-Chip Designs Milestone 2: Steven Beigelmacher Gall Gotfried

Scheduling Parallel Task

Created by: Ian Osborn. Possibilities Of Movie Maker Windows Movie Maker allows users to organize and add effects to media clips that ordinarily would.

Getting Started with Eclipse Sandeep Pasuparthy. What’s Eclipse? It is a free software / open source platform- independent software framework for delivering.

TrendReader Standard 2 This generation of TrendReader Standard software utilizes the more familiar Windows format (“tree”) views of functions and file.

Principles of Programming Chapter 1: Introduction  In this chapter you will learn about:  Overview of Computer Component  Overview of Programming 

Advanced Digital Circuits ECET 146 Week 3 Professor Iskandar Hack ET 221B,

Systems Analysis – Analyzing Requirements.  Analyzing requirement stage identifies user information needs and new systems requirements  IS dev team.

User Group Meeting Dashboard Features Products Customers Vendors Quick Find Menu/Search Shortcuts Popups Key Performance Indicators Tasks.

ISE. Tatjana Petrovic 249/982/22 ISE software tools ISE is Xilinx software design tools that concentrate on delivering you the most productivity available.

C HU H AI C OLLEGE O F H IGHER E DUCATION D EPARTMENT O F C OMPUTER S CIENCE Preparation of Final Year Project Report Bachelor of Science in Computer Science.

Introduction to Visual Basic. Quick Links Windows Application Programming Event-Driven Application Becoming familiar with VB Control Objects Saving and.

A Spring 2005 CS 426 Senior Project By Group 15 John Studebaker, Justin Gerthoffer, David Colborne CSE Dept., University of Nevada, Reno Advisors (CSE.

Chapter 2: Beginning the Problem-Solving Process

1 Scheduling CEG 4131 Computer Architecture III Miodrag Bolic Slides developed by Dr. Hesham El-Rewini Copyright Hesham El-Rewini.

Introduction With TimeCard users can tag SharePoint events with information that converts them into time sheets. This way they can report.

Computer Science and Engineering Parallel and Distributed Processing CSE 8380 March 01, 2005 Session 14.

DLS on Star (Single-level tree) Networks Background: A simple network model for DLS is the star network with a master-worker platform. It consists of a.

ASC2003 (July 15,2003)1 Uniformly Distributed Sampling: An Exact Algorithm for GA’s Initial Population in A Tree Graph H. S.

Teaching system for advanced statistics I. Nagy FD ČVUT, Prague J. Homolová FD ČVUT, Prague E. Suzdaleva ÚTIA AV ČR,

Computer Science and Engineering Parallel and Distributed Processing CSE 8380 March 08, 2005 Session 16.

ENHANCED MONITORING TOOL PROJECT Project Presentation By: David Nasi & Amitay Svetlit Supervisor: Oved Itzhak Software Systems Lab Department of Electrical.

Practice and Evaluation. Practice Develop a java class called: SumCalculator.java which computes a sum of all integer from 1 to 100 and displays the result.

1 Nasser Alsaedi. The ultimate goal for any computer system design are reliable execution of task and on time delivery of service. To increase system.

Problem Statement: Users can get too busy at work or at home to check the current weather condition for sever weather. Many of the free weather software.

An Introduction to Designing and Executing Workflows with Taverna Aleksandra Pawlik materials by: Katy Wolstencroft University of Manchester.

Optimal Algorithms for Task Scheduling Implemented by Ala Al-Nawaiseh Khaled Mahmud.

Computer Science and Engineering Parallel and Distributed Processing CSE 8380 March 03, 2005 Session 15.

Renesas Technology America Inc. 1 M16C Seminars Lab 3 Creating Projects Using HEW4 14 March 2005 M16C Seminars Lab 3 Creating Projects Using HEW4 Last.

Privacy Preserving Payments in Credit Networks By: Moreno-Sanchez et al from Saarland University Presented By: Cody Watson Some Slides Borrowed From NDSS’15.

Problem Reduction So far we have considered search strategies for OR graph. In OR graph, several arcs indicate a variety of ways in which the original.

A Roadmap towards Machine Intelligence

This tutorial will talk you through a very basic workbench queueing simulation. The queueing system modelled is of customers entering an infinite capacity.

Project Setup and Execution For PMPlan Enterprise Presented by AlNik Solutions Copyright ©

C HU H AI C OLLEGE O F H IGHER E DUCATION D EPARTMENT O F C OMPUTER S CIENCE Preparation of Final Year Project Report Bachelor of Science in Computer Science.

Results Table of contents Table of contents Table of contents Section 1 Table of contentsTable of contents Opening the Results module.

Tieflow Workflow Toolkit -:By:- CSCI 6831 Capstone project Instructor: Dr. Kwok-Bun Yue Mentor: Mr. Scott Hetherington Midterm Presentation 04/15/2008.

Planning and Scheduling.  A job can be made up of a number of smaller tasks that can be completed by a number of different “processors.”  The processors.

WS7-1 ADM , Workshop 7, August 2005 Copyright  2005 MSC.Software Corporation WORKSHOP 7 DESIGN OF EXPERIMENTS.

An Introduction to Programming with C++1 Beginning the Problem- Solving Process Tutorial 2.

Text2PTO: Modernizing Patent Application Filing A Proposal for Submitting Text Applications to the USPTO.

HTBN Batches These slides are intended as a starting point for further discussion of how eTime might be extended to allow easier processing of HTBN data.

McGraw-Hill/Irwin The Interactive Computing Series © 2002 The McGraw-Hill Companies, Inc. All rights reserved. Microsoft Excel 2002 Using Macros Lesson.

Click through this presentation at your own pace.

Visualizing Prim’s MST Algorithm Used to Trace the Algorithm in Class

Theory of Computation Turing Machines.

Laura Bright David Maier Portland State University

Richard Anderson Autumn 2016 Lecture 7

Planning and Scheduling

Planning and Scheduling

Richard Anderson Winter 2019 Lecture 7

Some Graph Algorithms.

Richard Anderson Autumn 2015 Lecture 7

Richard Anderson Autumn 2019 Lecture 7

Presentation transcript:

The El-Rewini/Ali Scheduling of In-Forest Task Graph on Two Processors with Communication Project Presentation By David Y. Feinstein SMU - CSE 8388 Spring 2005 Instructor: Dr. H. El-Rewini April 25, 2005

Problem description  An in-forest task graph allows only one successor task for each node. Out-forest task graphs are reversed.  The Hu Scheduling algorithm (1961) of in/out-forest task graphs on arbitrary number of processor assumes no communication cost.  The El-Rewini/Ali algorithm (1994) was the first to achieve optimal scheduling of in/out-forest tasks graph with communication. It does limit the maximum number of processors to two.  The El-Rewini/Ali algorithm introduced the notion of Augmental Graph to compensate for the communication cost. The Augmental Graph can be scheduled using the Hu algorithm followed by the SwapAll check to verify or correct the resulting schedule.

My proposed solution  This project provided an advanced framework to simulate both Hu algorithm (on arbitrary number of processors) and the El-Rewini/Ali algorithm.  User can create, edit and store task graphs using node and successor arc editors.  After the required algorithm is selected, the user can animate the process of the GANTT diagram creation.  Augmental graph creation is show in the detailed algorithm output memo control. Sibling analysis results can be seen graphically.  My project currently supports only the In-Forest – it can be easily extended for Out-Forest task graphs.

Assumptions Execution time of all tasks takes is one unit of time. Communication (in the El-Rewini/Ali algorithm) tasks one unit of time. Since the program currently supports only In-Forest task graphs, all successor arcs are pointing downward. (Arrow heads not shown for graph clarity) The file extension name for this project is “*.eas” (for El-Rewini Ali Scheduling)

Overview

In-Forest Task Graph Creation The user first generates the empty graph, after setting the number of levels. Use Node Insert/Delete set the proper number of nodes Use the successor editor to set the successor

Here is a sample of graph (also stored in “test1.eas”)

The Default Process Setup Selection is the El-Rewini/Ali optimal algorithm with Communication. The El-Rewini/Ali Algorithm is the default. Number of processor is automatically set to 2 Must have a created graph (or a graph file *.eas open).

The Hu Algorithm is Limited to No Communication When selecting the “Hu Only” process, the Hu algorithm will run without communication. You can set any number of processors. Users can simulate homework 4 Problem 10-1 (using file “test2.eas”)

The Augmental Graph Creation is an Essential Step of El-Rewini/Ali Algorithm The Augmental Graph process performs sibling analysis in the algorithm output screen. The use can select the “Show Sibling in Graph” or “Show Augmental Graph.

This partial display shows the detail algorithm output box

Other portions of the display show the Sibling Analysis and the Graph creation process.

Selecting “Show Sibling in Graph”

Selecting “Show Augmental Graph”

Notice what happens if we do not select “minimize arc crossing”. The result is still correct…

Hu Scheduling Control You can use animation or single step. The Hu Scheduling animation can be shown on the GANTT diagram or even on the graph itself. For El-Rewini/Ali, the Hu process works on the Augmental Graph.

During the Hu animation, Done nodes are in read, Ready nodes are shown in Green.

Gantt Diagram and program output listing details

The Final process in the El-Rewini/Ali is checking the resulting Hu Schedule. The program performs the SwapAll operation if required. Homework 4 problem 10-7 can be verified with the file test2.eas

File “test1.eas” required one swapall operation at time-6

Conclusions The El-Rewini/Ali Augmental Graph was a critical and elegant inventive step in order to solve the problem of scheduling in-forest task graphs with communication. Computation cost for the Augmental Graph creation are relatively minimal. When using the Hu Only (without communications - not show on the slides but can be run on the program) – we quickly reach a point where adding processors do not increase the output.