BOINC Berkeley Open Infrastructure for Network Computing An open-source middleware system for volunteer and grid computing (much of the images and text.

Slides:



Advertisements
Similar presentations
4/2/2002HEP Globus Testing Request - Jae Yu x Participating in Globus Test-bed Activity for DØGrid UTA HEP group is playing a leading role in establishing.
Advertisements

BOINC: A System for Public-Resource Computing and Storage David P. Anderson University of California, Berkeley.
Dinker Batra CLUSTERING Categories of Clusters. Dinker Batra Introduction A computer cluster is a group of linked computers, working together closely.
XSEDE 13 July 24, Galaxy Team: PSC Team:
David P. Anderson Space Sciences Laboratory University of California – Berkeley Volunteer Computing.
1 Distributed, Internet and Grid Computing. 2 Distributed Computing Current supercomputers are too expensive ASCI White (#1 in TOP500) costs more than.
Condor Overview Bill Hoagland. Condor Workload management system for compute-intensive jobs Harnesses collection of dedicated or non-dedicated hardware.
Cambodia-India Entrepreneurship Development Centre - : :.... :-:-
Scientific Computing on Smartphones David P. Anderson Space Sciences Lab University of California, Berkeley April 17, 2014.
Volunteer Computing and Hubs David P. Anderson Space Sciences Lab University of California, Berkeley HUBbub September 26, 2013.
Grid Toolkits Globus, Condor, BOINC, Xgrid Young Suk Moon.
Public-resource computing for CEPC Simulation Wenxiao Kan Computing Center/Institute of High Physics Energy Chinese Academic of Science CEPC2014 Scientific.
David Abarca, Instructor Del Mar College Computer Corner Use Your Computer To Help Science.
1 port BOSS on Wenjing Wu (IHEP-CC)
Open Science Grid For CI-Days Internet2: Fall Member Meeting, 2007 John McGee – OSG Engagement Manager Renaissance Computing Institute.
HTCondor and BOINC. › Berkeley Open Infrastructure for Network Computing › Grew out of began in 2002 › Middleware system for volunteer computing.
Volunteer Computing with BOINC David P. Anderson Space Sciences Laboratory University of California, Berkeley.
Scientific Computing in the Consumer Digital Infrastructure David P. Anderson Space Sciences Lab University of California, Berkeley The Austin Forum November.
Exa-Scale Volunteer Computing David P. Anderson Space Sciences Laboratory U.C. Berkeley.
Wenjing Wu Andrej Filipčič David Cameron Eric Lancon Claire Adam Bourdarios & others.
Wenjing Wu Computer Center, Institute of High Energy Physics Chinese Academy of Sciences, Beijing BOINC workshop 2013.
David Cameron Riccardo Bianchi Claire Adam Bourdarios Andrej Filipcic Eric Lançon Efrat Tal Hod Wenjing Wu on behalf of the ATLAS Collaboration CHEP 15,
07:44:46Service Oriented Cyberinfrastructure Lab, Introduction to BOINC By: Andrew J Younge
Evaluation of Agent Teamwork High Performance Distributed Computing Middleware. Solomon Lane Agent Teamwork Research Assistant October 2006 – March 2007.
BOINC.
National Science Foundation CI-TEAM Proposal: Blast on Condor How Will This Help [InstAbbrev]? Your Name Here Your Job Title Here Your Department Here.
Volunteer Computing with GPUs David P. Anderson Space Sciences Laboratory U.C. Berkeley.
and Citizen Cyber-Science David P. Anderson Space Sciences Laboratory U.C. Berkeley.
Ruth Pordes November 2004TeraGrid GIG Site Review1 TeraGrid and Open Science Grid Ruth Pordes, Fermilab representing the Open Science.
BOINC: Progress and Plans David P. Anderson Space Sciences Lab University of California, Berkeley BOINC:FAST August 2013.
David P. Anderson Space Sciences Laboratory University of California – Berkeley Designing Middleware for Volunteer Computing.
David P. Anderson Space Sciences Laboratory University of California – Berkeley Public and Grid Computing.
Introduction to Grids By: Fetahi Z. Wuhib [CSD2004-Team19]
TEMPLATE DESIGN © BOINC: Middleware for Volunteer Computing David P. Anderson Space Sciences Laboratory University of.
11/15/04PittGrid1 PittGrid: Campus-Wide Computing Environment Hassan Karimi School of Information Sciences Ralph Roskies Pittsburgh Supercomputing Center.
Globus and PlanetLab Resource Management Solutions Compared M. Ripeanu, M. Bowman, J. Chase, I. Foster, M. Milenkovic Presented by Dionysis Logothetis.
Grid Computing Unit I Introduction. Information anytime anywhere!!! support computation across administrative domains Generally  virtualizing computing.
Dr Jukka Klem CHEP06 1 Public Resource Computing at CERN – Philippe Defert, Markku Degerholm, Francois Grey, Jukka Klem, Juan Antonio.
1 Volunteer Computing at CERN past, present and future Ben Segal / CERN (describing the work of many people at CERN and elsewhere ) White Area lecture.
David P. Anderson Space Sciences Laboratory University of California – Berkeley Public Distributed Computing with BOINC.
BOINC: An Open Platform for Public-Resource Computing David P. Anderson Space Sciences Laboratory U.C. Berkeley.
David P. Anderson Space Sciences Laboratory University of California – Berkeley Public Distributed Computing with BOINC.
What is Protein Folding? Implications of Misfolding Computational Techniques Background image: Staphylococcal protein A, Z Domain (
SCI-BUS project Pre-kick-off meeting University of Westminster Centre for Parallel Computing Tamas Kiss, Stephen Winter, Gabor.
Centre for Parallel Computing Tamas Kiss Centre for Parallel Computing A Distributed Rendering Service Tamas Kiss Centre for Parallel Computing Research.
CernVM and Volunteer Computing Ivan D Reid Brunel University London Laurence Field CERN.
Volunteer Computing and BOINC Dr. David P. Anderson University of California, Berkeley Dec 3, 2010.
The Future of Volunteer Computing David P. Anderson U.C. Berkeley Space Sciences Lab UH CS Dept. March 22, 2007.
Emulating Volunteer Computing Scheduling Policies Dr. David P. Anderson University of California, Berkeley May 20, 2011.
Volunteer Computing: Involving the World in Science David P. Anderson U.C. Berkeley Space Sciences Lab February 16, 2007.
David P. Anderson Space Sciences Laboratory University of California – Berkeley Supercomputing with Personal Computers.
Volunteer Computing and Large-Scale Simulation David P. Anderson U.C. Berkeley Space Sciences Lab February 3, 2007.
Using volunteered resources for data-intensive computing and storage David Anderson Space Sciences Lab UC Berkeley 10 April 2012.
Volunteer Computing David P. Anderson U.C. Berkeley Space Sciences Lab Nov. 15, 2006.
Volunteer Computing with BOINC: a Tutorial David P. Anderson Space Sciences Laboratory University of California – Berkeley May 16, 2006.
Volunteer Computing David P. Anderson U.C. Berkeley Space Sciences Lab January 30, 2007.
An Overview of Volunteer Computing
Volunteer Computing and BOINC
Status of WLCG FCPPL project
University of California, Berkeley
Volunteer computing PC owners donate idle cycles to science projects
Volunteer Computing: SETI and Beyond David P
Volunteer Computing for Science Gateways
Designing a Runtime System for Volunteer Computing David P
Exa-Scale Volunteer Computing
Job Scheduling in a Grid Computing Environment
David P. Anderson Space Sciences Lab UC Berkeley LASER
The Global Status of Citizen Cyberscience
Grid Computing Colton Lewis.
US CMS Testbed.
Presentation transcript:

BOINC Berkeley Open Infrastructure for Network Computing An open-source middleware system for volunteer and grid computing (much of the images and text for this presentation are 'borrowed' under fair use and/or creative commons from boinc.berkeley.edu, wikipedia.com, westminster.ac.uk, cs.umd.edu, and cern.ch)

What is BOINC? Software that enables computers to put otherwise unused CPU and GPU cycles in to use for computationally intense projects, usually in science or math related fields Originally developed to manage the Designed to address security and malicious user issues that arose in prior to the development of Generalized to be an open platform that can run any sufficiently parallelized application

Why BOINC? Supercomputing on the cheap (all you need is a ~$5000 server to coordinate a project) Utilization of normally under-utilized computing resources (think about the computer labs on your campus) Spreading of awareness Sense of ownership and participation by the public

BOINC stats 287,945 volunteers 835,757 computers 50+ scientific projects open for the public to volunteer 24-hour average computational output: petaFLOPS Near infinite expandability

For comparison... the fastest supercomputer in use in the world is the Tianhe-1A peak performance of petaFLOPS Cost $88 million to build Costs $20 million annually to power and operate Requires a full time staff of 200 to operate

Volunteer computing vs Grid computing In Grid computing, organizations share resources. Any organization that is part of a grid computing effort can act as either an producer or consumer of resources. In Volunteer computing, volunteer individuals or organizations act as producers of resources, and only the coordinating organization may act as a consumer of resources.

How is BOINC used? Volunteer computing projects for science and math Virtual campus supercomputing centers (e.g univ. of Westminster in London) Desktop grids for business Integration with Condor to allow Globus- based grids to run jobs for BOINC projects (e.g. the Open Science Grid project)

Basic overview of BOINC jobs

1)Your PC gets a set of tasks from the project's scheduling server. Available tasks are constrained by the capabilities of your PC. 2)Your PC downloads executable and input files from the project's data server. If the project releases new versions of its applications, the executable files are downloaded automatically to your PC. 3)Your PC runs the application programs, producing output files. 4)Your PC uploads the output files to the data server. 5)Later, your PC reports the completed tasks to the scheduling server, and gets new tasks.

The BOINC client Notices tab: displays news from the projects in which you participate

The BOINC client Projects tab: overview of all your projects

The BOINC client Tasks tab: info on specific tasks within each project

The BOINC client Transfers tab: shows status of file transfers for all projects

The BOINC client Statistics tab: graphs your contribution to projects over time

The BOINC client Disk tab: shows BOINC's overall hard disk usage, and usage of each project

Using the BOINC client Selecting a project to volunteer for

Using the BOINC client Preferences dialog controls CPU, GPU, disk, network, and memory usage so that BOINC only consumes resources as directed by the user.

For each application, the BOINC core client creates a segment of shared memory that is used to pass messages between the core client and the application.

The BOINC client keeps track of how many CPU cycles are used in computation, and reports this information back to the project server. When at least 2 clients have reported completion of a task, the lower of their reported CPU cycles is used to calculate credit. BOINC's unit of credit, the Cobblestone (named after Jeff Cobb of is 1/200 day of CPU time on a reference computer that does 1,000 MFLOPS based on the Whetstone benchmark

If an application errors, its standard error is written to a file and transmitted back to the project server for analysis. If an application crashes or is aborted, a stack trace is written to standard error.

Generating work in BOINC (server side) Multiple jobs with different input files (file_1, file_2, etc)

Popular BOINC projects - analysis of radio telescope data, looking for patterns that may indicate the presence of extra-terrestrial intelligent life - simulates protein folding for biological research Climateprediction.net – forecasts weather and other climate conditions - search for spinning neutron stars (also called pulsars) using data from the LIGO and GEO gravitational wave detectors

Popular BOINC projects – 3d modeling of the Milky Way galaxy PrimeGrid – discovers large prime numbers World Community Grid – general purpose grid for humanitarian research - tries to determine the 3- dimensional shapes of proteins in research that may ultimately lead to finding cures for some major human diseases

Atom smasher?? It may not look like much, but this computer simulates high-energy particle collisions every day.

Test4theory A scientific application that uses BOINC as middleware to facilitate volunteer computing. A project of CERN that runs simulations of Large Hadron Collider experiments on volunteer machines. Computer simulations of high-energy particle collisions provide a detailed theoretical reference for the measurements performed at accelerators like the Large Hadron Collider (LHC), against which models of both known and 'new' physics can be tested, down to the level of individual particles.

By looking for discrepancies between the simulations and the data, we are searching for any sign of disagreement between the current theories and the physical universe. Ultimately, such a disagreement could lead us to the discovery of new phenomena, which may be associated with new fundamental principles of Nature Less spectacular discrepancies also help guide us towards the most accurate possible description of the Standard Model of Particle Physics and its phenomena - refining the simulations of the known physical laws, by pointing to areas where current simulations succeed and where they fail.

Top 10 test4theory participants … and me, ranked #451.

Campus Supercomputing Grid- University of Westminster Grid consists of ~1500 desktop PCs in labs all over campus Represents only about half of the PCs owned by the university Computing power equivalent to a £500,000 cluster procurement or supercomputer Cost to maintain is negligible... PCs on the grid are replaced and upgraded as normal from existing budgets. Environmental impact and energy usage are also negligible (for the same reasons)

Campus Supercomputing Grid- University of Westminster PCs simply need to have BOINC client installed. Settings can be locked down by campus IT admins so that only the campus' own BOINC projects can be run on university owned PCs. Project is coordinated from a single web server housed on campus. Any university department with projects that can benefit from supercomputing resources can use the grid.

University of Maryland: The Lattice Project The Lattice project incorporates the Globus toolkit and BOINC services with a higher level grid scheduler. Jobs are submitted to the grid, and are then assigned to either the pool of BOINC clients, a cluster master node, or an instance of Condor. Distribution of jobs is based on estimated run time and other resources needed. Since its inception, The Lattice Project has performed 21, CPU Years of computation.

Lattice Architecture

Applications that run on the Lattice Project

Lattice Project Resources (updated every 3 mins)