Volunteer Computing and Large-Scale Simulation David P. Anderson U.C. Berkeley Space Sciences Lab February 3, 2007.

Slides:



Advertisements
Similar presentations
BOINC: A System for Public-Resource Computing and Storage David P. Anderson University of California, Berkeley.
Advertisements

David P. Anderson Space Sciences Laboratory University of California – Berkeley Volunteer Computing.
Volunteer Computing David P. Anderson Space Sciences Lab U.C. Berkeley May 2, 2007.
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.
Public-resource computing for CEPC Simulation Wenxiao Kan Computing Center/Institute of High Physics Energy Chinese Academic of Science CEPC2014 Scientific.
Achievements and Opportunities in Volunteer Computing David P. Anderson Space Sciences Lab U.C. Berkeley 18 April 2008.
A Guided Tour of BOINC David P. Anderson Space Sciences Lab University of California, Berkeley TACC November 8, 2013.
Volunteer Computing David P. Anderson Space Sciences Lab U.C. Berkeley May 7, 2008.
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.
David P. Anderson Space Sciences Laboratory University of California – Berkeley Designing Middleware for Volunteer Computing.
Exa-Scale Volunteer Computing David P. Anderson Space Sciences Laboratory U.C. Berkeley.
Introduction to the BOINC software David P. Anderson Space Sciences Laboratory University of California, Berkeley.
David Cameron Claire Adam Bourdarios Andrej Filipcic Eric Lancon Wenjing Wu ATLAS Computing Jamboree, 3 December 2014 Volunteer Computing.
07:44:46Service Oriented Cyberinfrastructure Lab, Introduction to BOINC By: Andrew J Younge
BOINC.
Volunteer Computing with BOINC Dr. David P. Anderson University of California, Berkeley SC10 Nov. 14, 2010.
David P. Anderson Space Sciences Lab U.C. Berkeley Exa-Scale Volunteer Computing.
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.
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.
TEMPLATE DESIGN © BOINC: Middleware for Volunteer Computing David P. Anderson Space Sciences Laboratory University of.
Dr Jukka Klem CHEP06 1 Public Resource Computing at CERN – Philippe Defert, Markku Degerholm, Francois Grey, Jukka Klem, Juan Antonio.
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.
A Tour of Citizen Cyber-Science David P. Anderson Space Sciences Laboratory U.C. Berkeley.
Exa-Scale Volunteer Computing David P. Anderson Space Sciences Laboratory U.C. Berkeley.
Volunteer Computing Involving the World in Science David P. Anderson Space Sciences Lab U.C. Berkeley 13 December 2007.
Volunteer Computing David P. Anderson Space Sciences Lab U.C. Berkeley 14 Sept 2007.
Volunteer Computing: SETI and Beyond David P. Anderson University of California, Berkeley 7 June 2007.
Volunteer Computing and BOINC Dr. David P. Anderson University of California, Berkeley Dec 3, 2010.
Frontiers of Volunteer Computing David Anderson Space Sciences Lab UC Berkeley 30 Dec
The Future of Volunteer Computing David P. Anderson U.C. Berkeley Space Sciences Lab UH CS Dept. March 22, 2007.
Volunteer Computing in the Next Decade David Anderson Space Sciences Lab University of California, Berkeley 4 May 2012.
Volunteer Computing in Biology David P. Anderson Space Sciences Lab U.C. Berkeley 10 Sept 2007.
Emulating Volunteer Computing Scheduling Policies Dr. David P. Anderson University of California, Berkeley May 20, 2011.
David P. Anderson Space Sciences Laboratory University of California – Berkeley A Million Years of Computing.
Volunteer Computing: Involving the World in Science David P. Anderson U.C. Berkeley Space Sciences Lab February 16, 2007.
Volunteer Computing: the Ultimate Cloud Dr. David P. Anderson University of California, Berkeley Oct 19, 2010.
A Brief History of (CPU) Time -or- Ten Years of Multitude David P. Anderson Spaces Sciences Lab University of California, Berkeley 2 Sept 2010.
David P. Anderson Space Sciences Laboratory University of California – Berkeley Supercomputing with Personal Computers.
The Limits of Volunteer Computing Dr. David P. Anderson University of California, Berkeley March 20, 2011.
All the computers in the world (~1 billion) BOINC: high-level goal Computational science biology, medicine Earth sciences, physics, astronomy, math, A.I.,...
Volunteer Computing Involving the World in Science David P. Anderson Space Sciences Lab U.C. Berkeley 13 December 2007.
Using volunteered resources for data-intensive computing and storage David Anderson Space Sciences Lab UC Berkeley 10 April 2012.
Technology for Citizen Cyberscience Dr. David P. Anderson University of California, Berkeley May 2011.
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.
Frontiers of Volunteer Computing David Anderson Space Sciences Lab UC Berkeley 28 Nov
Volunteer Computing David P. Anderson U.C. Berkeley Space Sciences Lab January 30, 2007.
An Overview of Volunteer Computing
A Brief History of BOINC
Volunteer Computing and BOINC
University of California, Berkeley
Building a Global Brain David P. Anderson U. C
Volunteer computing PC owners donate idle cycles to science projects
Volunteer Computing: Planting the Flag David P
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.
Ivan Reid (Brunel University London/CMS)
Presentation transcript:

Volunteer Computing and Large-Scale Simulation David P. Anderson U.C. Berkeley Space Sciences Lab February 3, 2007

Volunteer computing ● Distributed computing using volunteered computers (mostly privately owned) ● Early projects ( ) – GIMPS – distributed.net – – ● Different from “Peer-to-Peer computing”, “Grid computing”

Volunteer computing is hard ● Resources are anonymous, untrusted – overclockers – hackers ● Resources are very – heterogeneous – intermittent – high-churn ● Incentives are needed to attract, retain volunteers – most important one: credit for work done (Middleware can address all these issues)

Why is it interesting? ● More/cheaper computing power than other paradigms – Currently ~1 billion PCs – Consumer products have economy-of-scale advantage ● What determines how much a scientist gets? – public appeal of research – public relations activities – good web site, message boards, graphics ● Scientific education/outreach – Public learns about research – Scientists publicize their research

Computing power and cost ● ~450,000 active computers ● Hardware cost: ~ $1/GFLOP (save 1000x) ● Future: game consoles (e.g. Playstation 3) – ExaFLOP computing feasible

BOINC ● Software platform for volunteer computing – Open-source (LGPL), NSF-funded – Supports wide range of application characteristics – Encourages multi-project participation ● Goal: create a new computing paradigm (not just a static set of projects) Volunteers Projects (constantly changing)

Some BOINC-based projects ● Climateprediction.net – Oxford; global climate change study ● – LIGO; gravitational wave astronomy ● – U. Washington; protein study ● – U.C. Berkeley; SETI ● – CERN; accelerator simulation ● – STI, U. of Geneva; malaria epidemiology ●...and about 30 others

Attachment and resource share Projects Hosts 60 % 40 %

BOINC communication structure client scheduler I’m a Windows/x86 computer with 512 MB RAM and 20GB free disk. I do 2.5 GFLOPS and 4.2 MIPS. I’m running 35% of the time, and your project gets 50% of that. I need enough work to last for 3 days Queue of jobs: RAM/disk/FLOPS requirements, soft deadline OK, here are three jobs. Job 1 has application files A,B,C, input files C,D,E and output file F...

Replicated computing host project The answer is 42. It took me 10 GFLOPs to compute. What if the host is wrong (or lying)? host project 778, 10 GFLOPs 39.8, 10.1 GFLOPs 42.03, 11 GFLOPs

Simulations and replications ● Most simulations are unstable – wildly different answers on different platforms – “fuzzy comparison” doesn’t work ● Homogeneous replication: send replicas to numerically equivalent hosts, demand identical answers ● Check that answer is internally consistent and/or that it is consistent with ensemble results

Application types ● Standard – C/C++, slight mods for file I/O, checkpointing – graphics: supply OpenGL render function ● FORTRAN ● Legacy (no source code) – BOINC-supplied “wrapper” ● Compound applications – master, workers ● Scripted – actual program is contained in input file or command line

Work flow work generator (creates stream or batches of jobs) assimilator (handles correst result) validator (compares replicas, selects “correct” result) BOINC

What apps are good for volunteer computing? ● Needs a lot of computer time – 30 days: don’t bother – 10,000 years: no problem ● Lots of independent tasks (or sequence of batches of independent tasks) – Simulations: perturb initial conditions ● Data/compute ratio fairly low (< 1 GB per day of computing) ● Public appeal

Creating a BOINC-based project ● Port or develop application – always Windows; maybe Mac OS X; maybe Linux ● Get a server – hardware: lots of RAM, storage; ~$5K – software: Linux, MySQL, Apache, BOINC ● Develop work-flow components ● Develop web site ● Testing ● Publicity ● Message boards and customer support

Organizational alternatives ● Projects ● Meta-projects – World Community Grid – UC Berkeley? – NIH? science sys admin web design PR, support science sys admin web design PR, support science host

Conclusion ● Volunteer computing – cheap supercomputing – connect to the public ● Well-suited to large-scale simulation ● BOINC: open-source platform for volunteer computing –