Deploying Research in the Real World: The OurGrid Experience

Slides:



Advertisements
Similar presentations
Buffers & Spoolers J L Martin Think about it… All I/O is relatively slow. For most of us, input by typing is painfully slow. From the CPUs point.
Advertisements

1 Bogotá, EELA-2 1 st Conference, On the Co-existence of Service and Opportunistic Grids Francisco Brasileiro Universidade Federal.
MyGrid: A User-Centric Approach for Grid Computing Walfredo Cirne Universidade Federal da Paraíba.
The OurGrid Project Walfredo Cirne Universidade Federal de Campina Grande.
The OurGrid Project Walfredo Cirne Universidade Federal de Campina Grande.
Chapter 3 Operating Systems. Chapter 3 Operating Systems 3.1 The Evolution of Operating Systems 3.1 The Evolution of Operating Systems 3.2 Operating System.
Running Thor over MyGrid Walfredo Cirne Universidade Federal de Campina Grande.
MCell Usage Scenario Project #7 CSE 260 UCSD Nadya Williams
Chapter 1 Introduction 1.1A Brief Overview - Parallel Databases and Grid Databases 1.2Parallel Query Processing: Motivations 1.3Parallel Query Processing:
The many-core architecture 1. The System One clock Scheduler (ideal) distributes tasks to the Cores according to a task map Cores 256 simple RISC Cores,
Labs of The World, Unite!!! Walfredo Cirne Universidade Federal de Campina Grande.
Cambodia-India Entrepreneurship Development Centre - : :.... :-:-
Client-Server Computing in Mobile Environments
Massive Ray Tracing in Fusion Plasmas on EGEE J.L. Vázquez-Poletti, E. Huedo, R.S. Montero and I.M. Llorente Distributed Systems Architecture Group Universidad.
MobSched: An Optimizable Scheduler for Mobile Cloud Computing S. SindiaS. GaoB. Black A.LimV. D. AgrawalP. Agrawal Auburn University, Auburn, AL 45 th.
Connecting OurGrid & GridSAM A Short Overview. Content Goals OurGrid: architecture overview OurGrid: short overview GridSAM: short overview GridSAM: example.
WP9 Resource Management Current status and plans for future Juliusz Pukacki Krzysztof Kurowski Poznan Supercomputing.
E-science grid facility for Europe and Latin America OurGrid E2GRIS1 Rafael Silva Universidade Federal de Campina.
Young Suk Moon Chair: Dr. Hans-Peter Bischof Reader: Dr. Gregor von Laszewski Observer: Dr. Minseok Kwon 1.
Multicriteria Driven Resource Management Strategies in GRMS Krzysztof Kurowski, Jarek Nabrzyski, Ariel Oleksiak, Juliusz Pukacki Poznan Supercomputing.
Wenjing Wu Andrej Filipčič David Cameron Eric Lancon Claire Adam Bourdarios & others.
Sep 21, 20101/14 LSST Simulations on OSG Sep 21, 2010 Gabriele Garzoglio for the OSG Task Force on LSST Computing Division, Fermilab Overview OSG Engagement.
A Survey of Distributed Task Schedulers Kei Takahashi (M1)
OurGrid: A Simple Solution for Running Bag-of-Tasks Applications on Grids Marcelo Meira, Walfredo Cirne (marcelo, Universidade.
1 Distributed Energy-Efficient Scheduling for Data-Intensive Applications with Deadline Constraints on Data Grids Cong Liu and Xiao Qin Auburn University.
Grid Workload Management Massimo Sgaravatto INFN Padova.
EFFECTIVE LOAD-BALANCING VIA MIGRATION AND REPLICATION IN SPATIAL GRIDS ANIRBAN MONDAL KAZUO GODA MASARU KITSUREGAWA INSTITUTE OF INDUSTRIAL SCIENCE UNIVERSITY.
Alexandre Duarte Gustavo Wagner Francisco Brasileiro Walfredo Cirne Multi-Environment Software Testing on the Grid Universidade Federal de Campina Grande.
JEMMA: an open platform for a connected Smart Grid Gateway GRUPPO TELECOM ITALIA MAS2TERING Smart Grid Workshop Brussels, September Strategy &
1 Catania, 4 th EEGE User Forum/OGF 25, OurGrid integration with gLite based grids in EELA-2 Francisco Brasileiro Universidade.
E-science grid facility for Europe and Latin America E2GRIS1 Gustavo Miranda Teixeira Ricardo Silva Campos Laboratório de Fisiologia Computacional.
1 Bogotá, EELA-2 1 st Conference, The OurGrid Approach for Opportunistic Grid Computing Francisco Brasileiro Universidade Federal.
E-science grid facility for Europe and Latin America Bridging the High Performance Computing Gap with OurGrid Francisco Brasileiro Universidade.
Data Replication and Power Consumption in Data Grids Susan V. Vrbsky, Ming Lei, Karl Smith and Jeff Byrd Department of Computer Science The University.
NGMAST 2008 A Proactive and Distributed QoS Negotiation Approach for Heterogeneous environments Anis Zouari, Lucian Suciu, Jean Marie Bonnin, and Karine.
E-science grid facility for Europe and Latin America OurGrid and the co-existence with gLite Alexandre Duarte Universidade Federal de Campina.
E-infrastructure shared between Europe and Latin America Interoperability between EELA and OurGrid Alexandre Duarte CERN and UFCG 1 st.
PROP: A Scalable and Reliable P2P Assisted Proxy Streaming System Computer Science Department College of William and Mary Lei Guo, Songqing Chen, and Xiaodong.
Uppsala, April 12-16th 2010EGEE 5th User Forum1 A Business-Driven Cloudburst Scheduler for Bag-of-Task Applications Francisco Brasileiro, Ricardo Araújo,
Ensieea Rizwani An energy-efficient management mechanism for large-scale server clusters By: Zhenghua Xue, Dong, Ma, Fan, Mei 1.
E-infrastructure shared between Europe and Latin America Interoperability between EELA and OurGrid Alexandre Duarte CERN IT-GD EELA Project.
Name : Mamatha J M Seminar guide: Mr. Kemparaju. GRID COMPUTING.
Measuring Performance Based on slides by Henri Casanova.
Using volunteered resources for data-intensive computing and storage David Anderson Space Sciences Lab UC Berkeley 10 April 2012.
8 th International Desktop Grid Federation Workshop, Hannover, Germany, August 17 th, 2011 DEGISCO Desktop Grids for International Scientific Collaboration.
EGI-InSPIRE RI EGI-InSPIRE EGI-InSPIRE RI EGI solution for high throughput data analysis Peter Solagna EGI.eu Operations.
DIRAC: Workload Management System Garonne Vincent, Tsaregorodtsev Andrei, Centre de Physique des Particules de Marseille Stockes-rees Ian, University of.
BAHIR DAR UNIVERSITY Institute of technology Faculty of Computing Department of information technology Msc program Distributed Database Article Review.
CPU SCHEDULING.
Introduction to Load Balancing:
Example: Rapid Atmospheric Modeling System, ColoState U
and Alexandre Duarte OurGrid/EELA Interoperability Meeting
Walter Binder Giovanna Di Marzo Serugendo Jarle Hulaas
Chapter 3 Internet Applications and Network Programming
QianZhu, Liang Chen and Gagan Agrawal
Paul Pop, Petru Eles, Zebo Peng
Grid Computing.
WLCG Accounting Task Force Update Julia Andreeva CERN WLCG Workshop 08
CompChem VO: User experience using MPI
Interoperability & Standards
Liang Chen Advisor: Gagan Agrawal Computer Science & Engineering
Chapter 2: The Linux System Part 1
CPU scheduling decisions may take place when a process:
Simulation in a Distributed Computing Environment
Chapter 6: CPU Scheduling
GATES: A Grid-Based Middleware for Processing Distributed Data Streams
Prof. Leonardo Mostarda University of Camerino
ShareGrid: architettura e middleware
Extreme Programming.
Implementation of a small-scale desktop grid computing infrastructure in a commercial domain    
Presentation transcript:

Deploying Research in the Real World: The OurGrid Experience Walfredo Cirne walfredo@dsc.ufcg.edu.br Universidade Federal de Campina Grande

OurGrid Goals and Scope Provide a real solution to real people Focus on Bag-of-Tasks (BoT) applications Serve as basis for our research

OurGrid Results A way to automatically join a grid, sharing resources in a fair manner Human intervention is not needed Schedulers that achieve good performance with no information about the grid nor the application A grid solution for BoT applications that you can download and use

Bag-of-Tasks Applications Data mining Massive search (as search for crypto keys) Parameter sweeps Monte Carlo simulations Fractals (such as Mandelbrot) Image manipulation (such as tomography) And many others…

OurGrid Architecture ... Site Manager User Interface SWAN SWAN n 1 ... n User Interface SWAN SWAN MyGrid Sandboxing

An Example: Factoring with MyGrid init mg-services put $PROC ./Fat.class $PLAYPEN grid1 java Fat 3 18655 34789789798 output-$TASK collect mg-services get $PROC $PLAYPEN output-$TASK grid2 java Fat 18655 37307 34789789798 output-$TASK

MyGrid GUI

Network of Favors OurGrid forms peer-to-peer community in which peers are free to join It’s important to encourage collaboration within OurGrid (i.e., resource sharing) In file-sharing, most users freeride OurGrid uses the Network of Favor All peers maintain a local balance for all known peers Peers with greater balances have priority The emergent behavior of the system is that by donating more, you get more resources No additional infrastructure is needed

NoF at Work [1] B 60 D 45 * * A D C E B ConsumerQuery broker A B C D E ConsumerQuery (broadcast) ConsumerFavor ProviderFavorReport ProviderWorkRequest * * * = no idle resources now

NoF at Work [2] B 60 D 45 E * * A D C E B ConsumerQuery A B C D E * ConsumerQuery ProviderWorkRequest * broker broker * = no idle resources now

Free-rider Consumption Epsilon is the fraction of resources consumed by free-riders

Equity Among Collaborators

Scheduling with No Information Grid scheduling typically depends on information about the grid (e.g. machine speed and load) and the application (e.g. task size) However, getting good information is hard Can we schedule without information and deploy the system now? Work-queue with Replication Tasks are sent to idle processors When there are no more tasks, running tasks are replicated on idle processors The first replica to finish is the official execution Other replicas are cancelled

Work-queue with Replication 8000 experiments Experiments varied in grid heterogeneity application heterogeneity application granularity Performance summary:

WQR Overhead Obviously, the drawback in WQR is cycles wasted by the cancelled replicas Wasted cycles:

Data Aware Scheduling WQR achieves good performance for CPU-intensive BoT applications However, many important BoT applications are data-intensive These applications frequently reuse data During the same execution Between two successive executions Storage Affinity uses replication and just a bit of static information to achieve good scheduling for data intensive applications

Storage Affinity Results 3000 experiments Experiments varied in grid heterogeneity application heterogeneity application granularity Performance summary: Storage Affinity X-Suffrage WQR Average (seconds) 57.046 59.523 150.270 Standard Deviation 39.605 30.213 119.200

Making it Work for Real...

OurGrid Status OurGrid is open source (GPL) and is available at www.ourgrid.org We’ve had external contributions OurGrid latest version is 3.0.2 It contains the 9th version of MyGrid The Network of Favors is available since version 3.0 We’ve had around 150 downloads SWAN is in final test phase To be released with version 3.1, in the end of April 2005

PAUÁ: Deploying OurGrid OurGrid has been deployed in the PAUÁ community by HP Brazil PAUÁ requires members to identify themselves and submit an adhesion proposal This makes people all the more comfortable (especially without SWAN) PAUÁ currently encompasses 7 sites (and growing fast)

The Community Status: http://pauastatus.lsd.ufcg.edu.br

HIV research with OurGrid prevalent in Africa O prevalent in Europe and Americas ? HIV-1 N majority in the world M HIV-2 A B C D F G H J K 18% in Brazil B,c,F

HIV protease + Ritonavir RMSD Subtype F Subtype B

The HIV Research Grid 55 machines in 6 administrative domains in the US and Brazil Task = 3.3 MB input, 1 MB output, 4 to 33 minutes of dedicated execution Ran 60 tasks in 38 minutes Speed-up is 29.2 for 55 machines Considering an 18.5-minute average machine

Conclusions We have an easy-to-join grid solution for Bag-of-Tasks applications working today Real users provide invaluable feedback for systems research Delivering results to real users is really cool! :-)

Questions?

Thank you! Merci! Danke! Grazie! Gracias! Obrigado! More at www.ourgrid.org