Berkeley Open Infrastructure for Network Computing

From Wikipedia, the free encyclopedia

  (Redirected from Boinc)
Jump to: navigation, search
BOINC
BOINC's current logo
BOINC's old logo
Current (top) and former (bottom) BOINC logos
Developer(s) University of California, Berkeley
Stable release 6.6.36 / 2009-06-10; 29 days ago
Operating system Cross-platform
Type Grid computing
License GNU Lesser General Public License
Website http://boinc.berkeley.edu/

The Berkeley Open Infrastructure for Network Computing (BOINC) is a non-commercial middleware system for volunteer and grid computing. It was originally developed to support the SETI@home project before it became useful as a platform for other distributed applications in areas as diverse as mathematics, medicine, molecular biology, climatology, and astrophysics. The intent of BOINC is to make it possible for researchers to tap into the enormous processing power of personal computers around the world.

BOINC has been developed by a team based at the Space Sciences Laboratory at the University of California, Berkeley led by David Anderson, who also leads SETI@home. As a "quasi-supercomputing" platform, BOINC has about 570,000 active computers (hosts) worldwide processing on average 1.9 PetaFLOPS as of July 2009,[1] which tops the processing power of the current fastest supercomputer system (IBM Roadrunner, with a sustained processing rate of 1.026 PFLOPS). BOINC is funded by the National Science Foundation through awards SCI/0221529,[2] SCI/0438443[3] and SCI/0721124.[4]

The framework is supported by various operating systems, including Microsoft Windows, Mac OS X and various Unix-like systems including Linux and FreeBSD. BOINC is free software which is released under the GNU Lesser General Public License.

Contents

[edit] History

BOINC was originally developed to manage the SETI@home project.

The original SETI client was a non-BOINC software exclusively for SETI@home. Being one of the first volunteer grid computing projects, it was not designed with a high level of security. Some participants in the project attempted to cheat the project to gain "credits"; while some others submitted entirely falsified work. BOINC was designed, in part, to combat these security breaches.[5]

The BOINC project started in February 2002 and the first version was released on 10 April 2002. The first BOINC-based project was Predictor@home launched on 9 June 2004.

[edit] Design and structure

BOINC is designed to be a free structure for anyone wishing to start a volunteer computing project. Most BOINC projects are nonprofit and rely heavily, if not completely, on volunteers.[citation needed]

In essence BOINC is software that can use the unused CPU and GPU cycles on a computer to do scientific computing—what one individual doesn't use of his/her computer, BOINC uses. In late 2008, BOINC's official website[6] announced that NVIDIA (a leading GPU manufacturer) had developed a system called CUDA that uses GPUs for scientific computing. With NVIDIA's assistance, some BOINC-based projects now have applications that run on NVIDIA GPUs using CUDA. These applications run from 2X to 10X faster than the former CPU-only versions.

BOINC consists of a server system and client software that communicate with each other to distribute, process, and return workunits.

[edit] User interfaces

BOINC Manager icon - Berkeley Open Infrastructure for Network Computing (BOINC), originally developed to support SETI@home, became useful as a platform for several distributed applications in areas as diverse as mathematics, medicine, molecular biology, climatology, and astrophysics.[7]

BOINC can be controlled remotely by Remote Procedure Calls, from the command line, and from the BOINC Account Manager.

BOINC Manager currently has two 'views': the Advanced View and the Simplified GUI. The Grid View was removed in the 6.6.x clients as it was redundant.

The appearance (skin) of the Simplified GUI is user-customizable, in that users can create their own designs.

[edit] Account managers

The account manager concept was conceived and developed jointly by GridRepublic and BOINC. Current account managers include:

The account managers are the unifying points for all the BOINC projects. They create a site that one can find and manage project settings with a single log in and password.

[edit] Credit system

The BOINC Credit System is designed to avoid cheating by validating results before granting credit.

  • A credit management system helps to ensure that users are returning results which are both scientifically and statistically accurate.
  • Online distributed computing is almost entirely a volunteer endeavor. For this reason projects are dependent on a complicated and variable mix of new users, long-term users, and retiring users.
  • There is no single generic reason why someone chooses to donate his or her computing resources to any given project.

[edit] Projects using BOINC framework

[edit] See also

[edit] References

[edit] External links

Personal tools