2008-05-14 Scientific Workflow Scheduling in Computational Grids Report: Wei-Cheng Lee 8th Grid Computing Conference IEEE 2007 – Planning, Reservation,

Slides:



Advertisements
Similar presentations
Hadi Goudarzi and Massoud Pedram
Advertisements

Pegasus on the Virtual Grid: A Case Study of Workflow Planning over Captive Resources Yang-Suk Kee, Eun-Kyu Byun, Ewa Deelman, Kran Vahi, Jin-Soo Kim Oracle.
CS 443 Advanced OS Fabián E. Bustamante, Spring 2005 Resource Containers: A new Facility for Resource Management in Server Systems G. Banga, P. Druschel,
A Dynamic World, what can Grids do for Multi-Core computing? Daniel Goodman, Anne Trefethen and Douglas Creager
Silberschatz, Galvin and Gagne  2002 Modified for CSCI 399, Royden, Operating System Concepts Operating Systems Lecture 19 Scheduling IV.
All Hands Meeting, 2006 Title: Grid Workflow Scheduling in WOSE (Workflow Optimisation Services for e- Science Applications) Authors: Yash Patel, Andrew.
Performance-responsive Middleware for Grid Computing Dr Stephen Jarvis High Performance Systems Group University of Warwick, UK High Performance Systems.
Resource Management of Grid Computing
Data Management in Cloud Workflow Systems Dong Yuan Faculty of Information and Communication Technology Swinburne University of Technology.
Universität Dortmund Robotics Research Institute Information Technology Section Grid Metaschedulers An Overview and Up-to-date Solutions Christian.
Example for Scheduling- Structures: Structured HPC Grids.
GridFlow: Workflow Management for Grid Computing Kavita Shinde.
A Grid Resource Broker Supporting Advance Reservations and Benchmark- Based Resource Selection Erik Elmroth and Johan Tordsson Reporter : S.Y.Chen.
QoS-constrained List Scheduling Heuristics for Parallel Applications on Grids 16-th Euromicro PDP Toulose, February 2008 QoS-CONSTRAINED LIST SCHEDULING.
Resource Manager for Grid with global job queue and with planning based on local schedules V.N.Kovalenko, E.I.Kovalenko, D.A.Koryagin, E.Z.Ljubimskii,
1 Optimizing Utility in Cloud Computing through Autonomic Workload Execution Reporter : Lin Kelly Date : 2010/11/24.
CONDOR DAGMan and Pegasus Selim Kalayci Florida International University 07/28/2009 Note: Slides are compiled from various TeraGrid Documentations.
Self-Organizing Agents for Grid Load Balancing Junwei Cao Fifth IEEE/ACM International Workshop on Grid Computing (GRID'04)
Research on cloud computing application in the peer-to-peer based video-on-demand systems Speaker : 吳靖緯 MA0G rd International Workshop.
Location-aware MapReduce in Virtual Cloud 2011 IEEE computer society International Conference on Parallel Processing Yifeng Geng1,2, Shimin Chen3, YongWei.
Online Spectrum Allocation for Cognitive Cellular Network Supporting Scalable Demands Jianfei Wang
Active Monitoring in GRID environments using Mobile Agent technology Orazio Tomarchio Andrea Calvagna Dipartimento di Ingegneria Informatica e delle Telecomunicazioni.
Implementing Processes and Process Management Brian Bershad.
EXPOSE GOOGLE APP ENGINE AS TASKTRACKER NODES AND DATA NODES.
Lecture 2 Process Concepts, Performance Measures and Evaluation Techniques.
Young Suk Moon Chair: Dr. Hans-Peter Bischof Reader: Dr. Gregor von Laszewski Observer: Dr. Minseok Kwon 1.
Combining the strengths of UMIST and The Victoria University of Manchester Utility Driven Adaptive Workflow Execution Kevin Lee School of Computer Science,
GRID’2012 Dubna July 19, 2012 Dependable Job-flow Dispatching and Scheduling in Virtual Organizations of Distributed Computing Environments Victor Toporkov.
Multicriteria Driven Resource Management Strategies in GRMS Krzysztof Kurowski, Jarek Nabrzyski, Ariel Oleksiak, Juliusz Pukacki Poznan Supercomputing.
임규찬. 1. Abstract 2. Introduction 3. Design Goals 4. Sample-Based Scheduling for Parallel Jobs 5. Implements.
A Survey of Distributed Task Schedulers Kei Takahashi (M1)
1 Distributed Energy-Efficient Scheduling for Data-Intensive Applications with Deadline Constraints on Data Grids Cong Liu and Xiao Qin Auburn University.
1 Using SchedFlow for Performance Evaluation of Workflow Applications Barton P. Miller University of Wisconsin Elisa Heyman Gustavo Martínez.
Condor Week 2005Optimizing Workflows on the Grid1 Optimizing workflow execution on the Grid Gaurang Mehta - Based on “Optimizing.
BOF: Megajobs Gracie: Grid Resource Virtualization and Customization Infrastructure How to execute hundreds of thousands tasks concurrently on distributed.
Combining the strengths of UMIST and The Victoria University of Manchester Adaptive Workflow Processing and Execution in Pegasus Kevin Lee School of Computer.
Performance evaluation of component-based software systems Seminar of Component Engineering course Rofideh hadighi 7 Jan 2010.
Job scheduling algorithm based on Berger model in cloud environment Advances in Engineering Software (2011) Baomin Xu,Chunyan Zhao,Enzhao Hua,Bin Hu 2013/1/251.
Advisor: Resource Selection 11/15/2007 Nick Trebon University of Chicago.
A Method for Transparent Admission Control and Request Scheduling in E-Commerce Web Sites S. Elnikety, E. Nahum, J. Tracey and W. Zwaenpoel Presented By.
Scheduling in HPC Resource Management System: Queuing vs. Planning Matthias Hovestadt, Odej Kao, Alex Keller, and Achim Streit 2003 Job Scheduling Strategies.
Chapter 2 Processes and Threads Introduction 2.2 Processes A Process is the execution of a Program More specifically… – A process is a program.
Service-oriented Resource Broker for QoS-Guaranteed in Grid Computing System Yichao Yang, Jin Wu, Lei Lang, Yanbo Zhou and Zhili Sun Centre for communication.
MROrder: Flexible Job Ordering Optimization for Online MapReduce Workloads School of Computer Engineering Nanyang Technological University 30 th Aug 2013.
GridLab Resource Management System (GRMS) Jarek Nabrzyski GridLab Project Coordinator Poznań Supercomputing and.
SPARRO Group, University of Regina 1 Portal Software: Browser-based Monte Carlo Zisis Papandreou University of Regina GlueX Collaboration Meeting JLab,
OPERATING SYSTEMS CS 3530 Summer 2014 Systems and Models Chapter 03.
Timeshared Parallel Machines Need resource management Need resource management Shrink and expand individual jobs to available sets of processors Shrink.
BalanceFlow: Controller load balancing for OpenFlow networks Hu, Yannan ; Wang, Wendong ; Gong, Xiangyang ; Que, Xirong ; Cheng, Shiduan Cloud Computing.
Scheduling MPI Workflow Applications on Computing Grids Juemin Zhang, Waleed Meleis, and David Kaeli Electrical and Computer Engineering Department, Northeastern.
Performance-responsive Scheduling for Grid Computing Dr Stephen Jarvis High Performance Systems Group University of Warwick, UK High Performance Systems.
Data Consolidation: A Task Scheduling and Data Migration Technique for Grid Networks Author: P. Kokkinos, K. Christodoulopoulos, A. Kretsis, and E. Varvarigos.
Network Weather Service. Introduction “NWS provides accurate forecasts of dynamically changing performance characteristics from a distributed set of metacomputing.
Net-Centric Software and Systems I/UCRC A Framework for QoS and Power Management for Mobile Devices in Service Clouds Project Lead: I-Ling Yen, Farokh.
Use of Performance Prediction Techniques for Grid Management Junwei Cao University of Warwick April 2002.
Joint Institute for Nuclear Research Synthesis of the simulation and monitoring processes for the data storage and big data processing development in physical.
Holding slide prior to starting show. Scheduling Parametric Jobs on the Grid Jonathan Giddy
Resource Allocation and Scheduling for Workflows Gurmeet Singh, Carl Kesselman, Ewa Deelman.
1 Performance Impact of Resource Provisioning on Workflows Gurmeet Singh, Carl Kesselman and Ewa Deelman Information Science Institute University of Southern.
Advanced Operating Systems CS6025 Spring 2016 Processes and Threads (Chapter 2)
Towards a High Performance Extensible Grid Architecture Klaus Krauter Muthucumaru Maheswaran {krauter,
OPERATING SYSTEMS CS 3502 Fall 2017
Duncan MacMichael & Galen Deal CSS 534 – Autumn 2016
Lecture Topics: 11/1 Processes Process Management
A Dynamic Critical Path Algorithm for Scheduling Scientific Workflow Applications on Global Grids e-Science IEEE 2007 Report: Wei-Cheng Lee
Auburn University COMP7500 Advanced Operating Systems I/O-Aware Load Balancing Techniques (2) Dr. Xiao Qin Auburn University.
A Simulator to Study Virtual Memory Manager Behavior
مهندسی مجدد فرآیندهای تجاری
Processes and Process Management
Resource and Service Management on the Grid
Presentation transcript:

Scientific Workflow Scheduling in Computational Grids Report: Wei-Cheng Lee 8th Grid Computing Conference IEEE 2007 – Planning, Reservation, and Data/Network-Awareness

Abstract The GRACCE scheduler applies advanced scheduling techniques, such as resource negotiation and reservation, data/network-aware scheduling and performance prediction in the resource allocation and execution planning process. To evaluate the scheduler, we have set up an experimental environment that models a computational grid in those aspects relevant to workflow scheduling. The results show the average performance improvement, using the GRACCE scheduler, is about 20% under high resource loads.

Outline WORKFLOW SCHEDULING IN COMPUTATIONAL GRIDS 2 Simulation and Performance Results 4 INTRODUCTION 31 RESOURCE ALLOCATION AND WORKFLOW EXECUTION PLANNING 33 Conclusions and Further Work 35

INTRODUCTION The very important issue in executing a scientific workflow in computational grids is how to map and schedule workflow tasks onto multiple distributed resources and handle task dependencies in a timely manner to deliver users’ expected performance. DAGMan Taverna Karajan Triana They require users to arrange resources for workflow tasks in advance and manually specify resource details. ASKALON Pegasus Gridbus The grid scheduling hierarchy is not taken into account, thus making assumptions that are too optimistic in computational grid environments.

INTRODUCTION GRACCE scheduler applies advanced scheduling techniques : Resource negotiation and reservation, Data/network-aware scheduling Performance prediction in the resource allocation Execution planning process The most of the performance improvement are accomplished through the reduction of the queue waiting time of workflow tasks on the resource’s local schedulers. (GRid Application Coordination, Collaboration and Execution)

WORKFLOW SCHEDULING IN COMPUTATIONAL GRIDS A workflow scheduler is an application-level scheduler whose goal is to improve the application performance, i.e. to complete the workflow execution as fast as possible.

A. Performance Analysis of Workflow Execution 1) Task Execution Time 2) Data Transfer Time 3) Queue Waiting Time B. Workflow Scheduling 1) Resource Allocation Strategies 2) Workflow-Orchestrated Co-Allocations 3) Network/Data-Aware Scheduling P C t1 t2 domain1 domain2 WORKFLOW SCHEDULING IN COMPUTATIONAL GRIDS Resource load The number of CPUs requested by the job The jobs currently running and queued in the local scheduler Just-in-time allocation Look-ahead allocation

C. GRACCE Scheduling Framework WORKFLOW SCHEDULING IN COMPUTATIONAL GRIDS

RESOURCE ALLOCATION AND WORKFLOW EXECUTION PLANNING The GRACCE Scheduler is a data/network-aware workflow scheduler : A. Resource Allocation for A Workflow Task : 1) Resource Discovery and Evaluation 2) Resource Negotiation and Reservation

RESOURCE ALLOCATION AND WORKFLOW EXECUTION PLANNING The resource negotiation and reservation process is thus a two-phase handshaking between the Allocator and the resource owner. Allocator Local Scheduler ResvRequest ResvResponse ResvAccept ResvConfirm This process can be represented by the following two formulas:

RESOURCE ALLOCATION AND WORKFLOW EXECUTION PLANNING B. Workflow Execution Planning To plan the workflow execution, two important parameters for each task are required : 1. The (estimated) execution time of the task on the candidate resource. 2. The network bandwidth between the target resources for its parent tasks and the candidate resource for the task. Using a Predictor to represent the prediction operations as follows:

RESOURCE ALLOCATION AND WORKFLOW EXECUTION PLANNING

RESOURCE ALLOCATION AND WORKFLOW EXECUTION PLANNING

SIMULATION AND PERFORMANCE EVALUATION A. The Simulation Environment 1) Simulation of Grid Resources and Local Schedulers

SIMULATION AND PERFORMANCE EVALUATION 2) Simulation of Job Execution A job execution is simulated using a timer thread: when the thread starts, the job starts; when it times out, the job completes. 3) Random Job Generator The job generator is able to maintain the average resource load at a specific value between 0.0 and 1.0. If the current resource load is less than the expected load, the job generator creates and submits jobs.

SIMULATION AND PERFORMANCE EVALUATION B. Performance Evaluation 1) Performance Evaluation of a 7-Task Workflow 2) Performance Evaluation of a 20-Task Workflow 3) Collect the execution times of 60 workflows that were generated by a random workflow generator.

) Performance Evaluation of a 7-Task Workflow SIMULATION AND PERFORMANCE EVALUATION

) Performance Evaluation of a 7-Task Workflow SIMULATION AND PERFORMANCE EVALUATION

) Performance Evaluation of a 7-Task Workflow SIMULATION AND PERFORMANCE EVALUATION

) Performance Evaluation of a 7-Task Workflow SIMULATION AND PERFORMANCE EVALUATION

) Performance Evaluation of a 7-Task Workflow SIMULATION AND PERFORMANCE EVALUATION

) Performance Evaluation of a 7-Task Workflow SIMULATION AND PERFORMANCE EVALUATION

) Performance Evaluation of a 20-Task Workflow SIMULATION AND PERFORMANCE EVALUATION

) Performance Evaluation of a 20-Task Workflow SIMULATION AND PERFORMANCE EVALUATION

) SIMULATION AND PERFORMANCE EVALUATION

SIMULATION AND PERFORMANCE EVALUATION

Wei-Cheng Lee