i o scheduling service for multi application clusters
play

I/O Scheduling Service for Multi-Application Clusters Adrien Lebre - PowerPoint PPT Presentation

Dec 01, 2023 •201 likes •1.08k views

I/O Scheduling Service for Multi-Application Clusters Adrien Lebre Adrien.Lebre@irisa.fr Guillaume Huard, Yves Denneulin { Adrien.Lebre,Guillaume.Huard,Yves.Denneulin } @imag.fr Laboratoire ID-IMAG (UMR 5132), Grenoble, France. BULL - HPC,

  1. I/O Scheduling Service for Multi-Application Clusters Adrien Lebre Adrien.Lebre@irisa.fr Guillaume Huard, Yves Denneulin { Adrien.Lebre,Guillaume.Huard,Yves.Denneulin } @imag.fr Laboratoire ID-IMAG (UMR 5132), Grenoble, France. BULL - HPC, ´ Echirolles, France. Przemyslaw Sowa sowa@icis.pcz.pl Institute of Computer and Information Sciences Czestochowa University of Technology, Poland. Adrien Lebre aIOLi - workshop Phenix - December 2006 � Bull-ID LIPS 2006 c

  2. Plan Part 1 - Parallel Input/Output and Clusters Part 2 - Controlling and Scheduling Multi-application I/O Part 3 - aIOLi, an Input/Output Scheduler for HPC Part 4 - Conclusion Adrien Lebre 2/26 aIOLi - workshop Phenix - December 2006 � Bull-ID LIPS 2006 c

  3. Plan Part 1 - Parallel Input/Output and Clusters 1 Introduction Context Parallel I/O 2 Parallel I/O and Clusters Available Solutions 3 Objectives Part 2 - Controlling and Scheduling Multi-application I/O Part 3 - aIOLi, an Input/Output Scheduler for HPC Part 4 - Conclusion Adrien Lebre 3/26 aIOLi - workshop Phenix - December 2006 � Bull-ID LIPS 2006 c

  4. Context Environment Clusters of SMPs Linux High Performance Computing Intensive I/O applications CPU bounded application ⇒ I/O bounded application Remote hard drive I/O Parallel I/O Handling concurrent accesses to a same resource (a file) Accesses: different in size, in offset Example: a parallel matrix product Adrien Lebre 4/26 aIOLi - workshop Phenix - December 2006 � Bull-ID LIPS 2006 c

  5. Parallel I/O - Example Parallel matrix product Specific parts to fetch according to the data distribution (columns/rows, BLOCK/BLOCK ...) Matrix A Matrix B Matrix C a b c ... ... 1 2 3 ... ... ? ? ? ... ... d e f ... ... 4 5 6 ... ... ? ? ? ... ... g h i ... ... 7 8 9 ... ... ? ? ? ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... Matrices are stored "row by row" in files Matrix A a b c ... ... d e f ... ... g h i ... ... ... ... ... ... Matrix B 1 2 3 ... ... 4 5 6 ... ... 7 8 9 ... ... ... ... ... ... Adrien Lebre 5/26 aIOLi - workshop Phenix - December 2006 � Bull-ID LIPS 2006 c

  6. Parallel I/O - Example Parallel matrix product Specific parts to fetch according to the data distribution (columns/rows, BLOCK/BLOCK ...) Matrix A Matrix B Matrix C 1 read (n) P 0 n read (1) a b c ... ... 1 2 3 ... ... ? ? ? ... ... d e f ... ... 4 5 6 ... ... ? ? ? ... ... g h i ... ... 7 8 9 ... ... ? ? ? ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... Matrices are stored "row by row" in files Matrix A a b c ... ... d e f ... ... g h i ... ... ... ... ... ... Matrix B 1 2 3 ... ... 4 5 6 ... ... 7 8 9 ... ... ... ... ... ... Adrien Lebre 5/26 aIOLi - workshop Phenix - December 2006 � Bull-ID LIPS 2006 c

  7. Parallel I/O - Example Parallel matrix product Specific parts to fetch according to the data distribution (columns/rows, BLOCK/BLOCK ...) Matrix A Matrix B Matrix C 1 read (n) P 0 n read (1) a b c ... ... 1 2 3 ... ... ? ? ? ... ... 1 read (n) d e f ... ... 4 5 6 ... ... ? ? ? ... ... P 1 n read (1) g h i ... ... 7 8 9 ... ... ? ? ? ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... Matrices are stored "row by row" in files Matrix A a b c ... ... d e f ... ... g h i ... ... ... ... ... ... Matrix B 1 2 3 ... ... 4 5 6 ... ... 7 8 9 ... ... ... ... ... ... Adrien Lebre 5/26 aIOLi - workshop Phenix - December 2006 � Bull-ID LIPS 2006 c

  8. Parallel I/O - Example Parallel matrix product Specific parts to fetch according to the data distribution (columns/rows, BLOCK/BLOCK ...) File decomposition: Lot of disjoint/contiguous requests at the same time ⇒ ”lethal” behaviour for I/O subsystem Matrix A Matrix B Matrix C 1 read (n) P 0 n read (1) a b c ... ... 1 2 3 ... ... ? ? ? ... ... 1 read (n) d e f ... ... 4 5 6 ... ... ? ? ? ... ... P 1 n read (1) g h i ... ... 7 8 9 ... ... ? ? ? ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... 1 read (n) P i ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... n read (1) Matrices are stored "row by row" in files 1 read (n) P p n read (1) Matrix A a b c ... ... d e f ... ... g h i ... ... ... ... ... ... Matrix B 1 2 3 ... ... 4 5 6 ... ... 7 8 9 ... ... ... ... ... ... n2 * read (n) n2 * n read (1) Adrien Lebre 5/26 aIOLi - workshop Phenix - December 2006 � Bull-ID LIPS 2006 c

  9. Parallel I/O - Example No defined order between requests ⇒ many disk head movements Matrix A Matrix B Matrix C P 0 a b c ... ... 1 2 3 ... ... ? ? ? ... ... d e f ... ... 4 5 6 ... ... ? ? ? ... ... P 1 g h i ... ... 7 8 9 ... ... ? ? ? ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... P i ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... Matrix are stored "row by row" in files P p Matrix A a b c ... ... d e f ... ... g h i ... ... ... ... ... ... SEEK SEEK SEEK SEEK time Adrien Lebre 5/26 aIOLi - workshop Phenix - December 2006 � Bull-ID LIPS 2006 c

  10. Parallel I/O - Example No defined order between requests ⇒ many disk head movements Matrix A Matrix B Matrix C P 0 a b c ... ... 1 2 3 ... ... ? ? ? ... ... d e f ... ... 4 5 6 ... ... ? ? ? ... ... P 1 g h i ... ... 7 8 9 ... ... ? ? ? ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... P i ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... Matrix are stored "row by row" in files P p Matrix A a b c ... ... d e f ... ... g h i ... ... ... ... ... ... SEEK One HDD provides the best performances when files are accessed in a sequential and contiguous way SEEK SEEK SEEK time Adrien Lebre 5/26 aIOLi - workshop Phenix - December 2006 � Bull-ID LIPS 2006 c

  11. Parallel I/O - Example No defined order between requests ⇒ many disk head movements Matrix A Matrix B Matrix C P 0 a b c ... ... 1 2 3 ... ... ? ? ? ... ... d e f ... ... 4 5 6 ... ... ? ? ? ... ... P 1 g h i ... ... 7 8 9 ... ... ? ? ? ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... P i ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... Matrix are stored "row by row" in files P p Matrix A a b c ... ... d e f ... ... g h i ... ... ... ... ... ... Matrix B 1 2 3 ... ... 4 5 6 ... ... 7 8 9 ... ... ... ... ... ... Similar behavior for the matrix B ”Random order” between requests The bigger the number of requests the bigger the potential number of seeks ⇒ performance degradation (bottlenecks) Adrien Lebre 5/26 aIOLi - workshop Phenix - December 2006 � Bull-ID LIPS 2006 c

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend

User-level scheduling Don Porter CSE 506 Context Multi-threaded application; more threads

User-level scheduling Don Porter CSE 506 Context Multi-threaded application; more threads

User-level scheduling Don Porter CSE 506 Context Multi-threaded application; more threads than CPUs Simple threading approach: Create a kernel thread for each application thread OS does all the scheduling work Simple

566 views • 23 slides
User-level scheduling OS does all the scheduling work Simple as that! Don Porter

User-level scheduling OS does all the scheduling work Simple as that! Don Porter

11/14/11 Context Multi-threaded application; more threads than CPUs Simple threading approach: Create a kernel thread for each application thread User-level scheduling OS does all the scheduling work Simple as

350 views • 6 slides
Efficient Wake-Up Scheduling for Efficient Wake-Up Scheduling for Multi-Core Systems Multi-Core

Efficient Wake-Up Scheduling for Efficient Wake-Up Scheduling for Multi-Core Systems Multi-Core

TH EDA Efficient Wake-Up Scheduling for Efficient Wake-Up Scheduling for Multi-Core Systems Multi-Core Systems Ming-Chao Lee*, Yiyu Shi**, Yu-Guan Chen*, Diana Marculescu***, Shih-Chieh Chang* * Dept. of CS, National Tsing Hua

338 views • 21 slides
Learning Scheduling Algorithms for Data Processing Clusters Hongzi Mao, Malte Schwarzkopf,

Learning Scheduling Algorithms for Data Processing Clusters Hongzi Mao, Malte Schwarzkopf,

Learning Scheduling Algorithms for Data Processing Clusters Hongzi Mao, Malte Schwarzkopf, Shaileshh Bojja Venkatakrishnan, Zili Meng, Mohammad Alizadeh Mot Motivation on Scheduling is a fundamental task in computer systems Cluster

601 views • 34 slides
Multi-criteria scheduling Denis Trystram with the help of P-F. Dutot, K. Rzadca and E. Saule

Multi-criteria scheduling Denis Trystram with the help of P-F. Dutot, K. Rzadca and E. Saule

Introduction Basics on classical scheduling multi-objective scheduling One step further Fairness Alternative approach Conclusion Multi-criteria scheduling Denis Trystram with the help of P-F. Dutot, K. Rzadca and E. Saule LIG, Grenoble

2.02k views • 82 slides
Predictability using Multiple Virtual Lanes in Modern Multi-Core InfiniBand Clusters Hari

Predictability using Multiple Virtual Lanes in Modern Multi-Core InfiniBand Clusters Hari

Improving Application Performance and Predictability using Multiple Virtual Lanes in Modern Multi-Core InfiniBand Clusters Hari Subramoni, Ping Lai, Sayantan Sur and Dhabhaleswar. K. Panda Department of Computer Science & Engineering The

551 views • 25 slides
Scheduling on clusters and grids Gr egory Mouni e, Yves Robert et Denis Trystram ID-IMAG 6

Scheduling on clusters and grids Gr egory Mouni e, Yves Robert et Denis Trystram ID-IMAG 6

Some basics on scheduling theory Taking into account Communications Grid : towards non-standard models Concluding Remarks Scheduling on clusters and grids Gr egory Mouni e, Yves Robert et Denis Trystram ID-IMAG 6 mars 2006

1.01k views • 58 slides
Scheduling Multi-Periodic Mixed-Criticality DAGs on Multi-Core Architectures Roberto MEDINA

Scheduling Multi-Periodic Mixed-Criticality DAGs on Multi-Core Architectures Roberto MEDINA

Scheduling Multi-Periodic Mixed-Criticality DAGs on Multi-Core Architectures Roberto MEDINA Etienne BORDE Laurent PAUTET December 13, 2018 1/28 Outline Research Context Problem Statement Scheduling MC-DAGs on multi-cores Case Study

799 views • 31 slides
Dynamic list scheduling of threads on clusters G. G. H. Cavalheiro, E. D. Benitez, D. S.

Dynamic list scheduling of threads on clusters G. G. H. Cavalheiro, E. D. Benitez, D. S.

Dynamic list scheduling of threads on clusters G. G. H. Cavalheiro, E. D. Benitez, D. S. Peranconi, E. Moschetta Universidade do Vale do Rio dos Sinos Programa Interdisciplinar de Ps Graduao em Computao Aplicada DSM 2006 Overview

459 views • 22 slides
Shared Clusters Jack Li , Calton Pu Yuan Chen , Vanish Talwar, Dejan Milojicic Georgia Institute

Shared Clusters Jack Li , Calton Pu Yuan Chen , Vanish Talwar, Dejan Milojicic Georgia Institute

Improving Preemptive Scheduling with Application-Transparent Checkpointing in Shared Clusters Jack Li , Calton Pu Yuan Chen , Vanish Talwar, Dejan Milojicic Georgia Institute of Technology Hewlett Packard Labs Shared Clusters for Big Data

522 views • 21 slides
Decentralized Dynamic Scheduling across Heterogeneous Multi core across Heterogeneous Multi

Decentralized Dynamic Scheduling across Heterogeneous Multi core across Heterogeneous Multi

Decentralized Dynamic Scheduling across Heterogeneous Multi core across Heterogeneous Multi core Desktop Grids J Jaehwan Lee , Pete Keleher, Alan Sussman h L P t K l h Al S Department of Computer Science University of Maryland Multi

357 views • 25 slides
Learning Scheduling Algorithms for Data Processing Clusters Aakhila Shaheen Introduction

Learning Scheduling Algorithms for Data Processing Clusters Aakhila Shaheen Introduction

Learning Scheduling Algorithms for Data Processing Clusters Aakhila Shaheen Introduction Cluster schedulers prioritize generality, ease of understanding over achieving ideal performance Efficient utilization of compute resources can save

475 views • 9 slides
Job-scheduling for Mul3-user MapReduce clusters Outline

Job-scheduling for Mul3-user MapReduce clusters Outline

Job-scheduling for Mul3-user MapReduce clusters Outline Background Hadoop On Demand Issues FAIR Scheduler Obstacles to Fair Sharing Background

127 views • 10 slides
Dynamic Scheduling for Work Agglomeration on Heterogeneous Clusters Jonathan Lifflander, G. Carl

Dynamic Scheduling for Work Agglomeration on Heterogeneous Clusters Jonathan Lifflander, G. Carl

Dynamic Scheduling for Work Agglomeration on Heterogeneous Clusters Jonathan Lifflander, G. Carl Evans, Anshu Arya, Laxmikant Kale University of Illinois Urbana-Champaign May 7, 2012 Work is overdecomposed in objects Fine-grain task

496 views • 22 slides
Designing Multi-Leader based Allgather Algorithms for Multi-core Clusters Krishna Kandalla, Hari

Designing Multi-Leader based Allgather Algorithms for Multi-core Clusters Krishna Kandalla, Hari

Designing Multi-Leader based Allgather Algorithms for Multi-core Clusters Krishna Kandalla, Hari Subramoni, Gopal Santhanaraman, Matthew Koop and Dhabaleswar. K. Panda Computer Science & Engineering Department The Ohio State University

644 views • 46 slides
Tarcil: Reconciling Scheduling Speed and Quality in Large Shared Clusters Christina Delimitrou 1 ,

Tarcil: Reconciling Scheduling Speed and Quality in Large Shared Clusters Christina Delimitrou 1 ,

Tarcil: Reconciling Scheduling Speed and Quality in Large Shared Clusters Christina Delimitrou 1 , Daniel Sanchez 2 and Christos Kozyrakis 1 1 Stanford University, 2 MIT SOCC August 27 th 2015 Executive Summary Goals of

958 views • 29 slides
MIT Presenters Prof. Kamal Youcef-T oumi Dr. Kate Anderson Mrs. Theresa Werth 1 PRESENTATION

MIT Presenters Prof. Kamal Youcef-T oumi Dr. Kate Anderson Mrs. Theresa Werth 1 PRESENTATION

MIT Presenters Prof. Kamal Youcef-T oumi Dr. Kate Anderson Mrs. Theresa Werth 1 PRESENTATION OUTLINE Program History 1. About MIT 2. Fellowship Overview and 3. Benefits Program Activities 4. Life at MIT/In Boston 5. How to Apply

557 views • 30 slides
Welwyn Components Ltd Microelectronics Why Use Microelectronics? Braking resistors/Heaters

Welwyn Components Ltd Microelectronics Why Use Microelectronics? Braking resistors/Heaters

Welwyn Components Ltd Microelectronics Why Use Microelectronics? Braking resistors/Heaters Power & Control High Power High Density & Reliability High Frequency Thick Film Circuit Thick Film Ink Systems Printing Firing -

397 views • 21 slides
LAC 4932 C-130H WARNING This document contains technical data whose export may be

LAC 4932 C-130H WARNING This document contains technical data whose export may be

LAC 4932 C-130H WARNING This document contains technical data whose export may be restricted by the Arms Export Control Act (Title 22, USC, Sec 2751, et seq) or the Export Administration Act of 1979 as amended, Title 50, USC, App 2401,

444 views • 15 slides
Modeling and Simulation of UAV Carrier Landings Gaurav Misra , Tianyu Gao , and Xiaoli Bai

Modeling and Simulation of UAV Carrier Landings Gaurav Misra , Tianyu Gao , and Xiaoli Bai

Modeling and Simulation of UAV Carrier Landings Gaurav Misra , Tianyu Gao , and Xiaoli Bai Rutgers, The State University of New Jersey, Piscataway, NJ, 08854 With UAVs promising capabilities to increase operation flexibility and

642 views • 21 slides
19 July 2018 Dear Parents/Guardians, This year the Rugby Presentation Night will take the form of

19 July 2018 Dear Parents/Guardians, This year the Rugby Presentation Night will take the form of

19 July 2018 Dear Parents/Guardians, This year the Rugby Presentation Night will take the form of a dinner on Friday 24 August, at the Bentleigh Club (33 Yawla St, Bentleigh), from 7 pm onwards. The dinner is for all rugby players and families

63 views • 4 slides
The Bat Mitzvah Prea reaknes kness s Hill Hills C s Cou ountr ntry C y Club ub War

The Bat Mitzvah Prea reaknes kness s Hill Hills C s Cou ountr ntry C y Club ub War

The Bat Mitzvah Prea reaknes kness s Hill Hills C s Cou ountr ntry C y Club ub War Warm & m & We Welcomi oming Exc Exclusi lusivit ity, Exc Excepti tion onal al Cuis Cuisin ine, Exq Exquis isit ite Servic

299 views • 12 slides
MOUTARDERIE.BE OUR PRODUCTS MUSTARDS Mustard manufacturer since the beginning of our

MOUTARDERIE.BE OUR PRODUCTS MUSTARDS Mustard manufacturer since the beginning of our

MUSTARDS SAUCES READY MEALS The flexibility of a small company with the quality of a big one MOUTARDERIE.BE OUR PRODUCTS MUSTARDS Mustard manufacturer since the beginning of our activities, we offer a 100% natural mustard, that we use in

326 views • 6 slides
Open Camera Phone App and Hold over Bar Code. Click the Link above to See Our Menu, Specials

Open Camera Phone App and Hold over Bar Code. Click the Link above to See Our Menu, Specials

Open Camera Phone App and Hold over Bar Code. Click the Link above to See Our Menu, Specials & Wine List. Enjoy! 1:32 PM Thu Jun 11 History . Simple QR Code Scanner a .. . GET ***** .K Simple QRC eReader Easy Simple for QR

393 views • 11 slides

More recommend