The Berkeley Open Infrastructure for Network Computing^[2] (BOINC, pronounced /bɔɪŋk/ – rhymes with "oink"^[3]) is an open-source middleware system for volunteer computing (a type of distributed computing).^[4] Developed originally to support SETI@home,^[5] it became the platform for many other applications in areas as diverse as medicine, molecular biology, mathematics, linguistics, climatology, environmental science, and astrophysics, among others.^[6] The purpose of BOINC is to enable researchers to utilize processing resources of personal computers and other devices around the world.

BOINC development began with a group based at the Space Sciences Laboratory (SSL) at the University of California, Berkeley, and led by David P. Anderson, who also led SETI@home. As a high-performance volunteer computing platform, BOINC brings together 34,236 active participants employing 136,341 active computers (hosts) worldwide, processing daily on average 20.164 PetaFLOPS as of 16 November 2021^[update][7] (it would be the 21st largest processing capability in the world compared with an individual supercomputer).^[8] The National Science Foundation (NSF) funds BOINC through awards SCI/0221529,^[9] SCI/0438443^[10] and SCI/0721124.^[11] Guinness World Records ranks BOINC as the largest computing grid in the world.^[12]

BOINC code runs on various operating systems, including Microsoft Windows, macOS, Android,^[13] Linux, and FreeBSD.^[14] BOINC is free software released under the terms of the GNU Lesser General Public License (LGPL).

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

The original SETI client was a non-BOINC software exclusively for SETI@home. It was one of the first volunteer computing projects, and not designed with a high level of security. As a result, some participants in the project attempted to cheat the project to gain "credits", while others submitted entirely falsified work. BOINC was designed, in part, to combat these security breaches.^[15]

The BOINC project started in February 2002, and its first version was released on April 10, 2002. The first BOINC-based project was Predictor@home, launched on June 9, 2004. In 2009, AQUA@home deployed multi-threaded CPU applications for the first time,^[16] followed by the first OpenCL application in 2010.

As of 15 August 2022, there are 33 projects on the official list.^[17] There are also, however, BOINC projects not included on the official list. Each year, an international BOINC Workshop is hosted to increase collaboration among project administrators. In 2021, the workshop was hosted virtually.^[18]

While not affiliated with BOINC officially, there have been several independent projects that reward BOINC users for their participation, including Charity Engine (sweepstakes based on processing power with prizes funded by private entities who purchase computational time of CE users), Bitcoin Utopia (now defunct), and Gridcoin (a blockchain which mints coins based on processing power).

BOINC is software that can exploit the unused CPU and GPU cycles on computer hardware to perform scientific computing. In 2008, BOINC's website announced that Nvidia had developed a language called CUDA that uses GPUs for scientific computing. With NVIDIA's assistance, several BOINC-based projects (e.g., MilkyWay@home. SETI@home) developed applications that run on NVIDIA GPUs using CUDA. BOINC added support for the ATI/AMD family of GPUs in October 2009. The GPU applications run from 2 to 10 times faster than the former CPU-only versions. GPU support (via OpenCL) was added for computers using macOS with AMD Radeon graphic cards, with the current BOINC client supporting OpenCL on Windows, Linux, and macOS. GPU support is also provided for Intel GPUs.^[19]

BOINC consists of a server system and client software that communicate to process and distribute work units and return results.

BOINC is used by many groups and individuals. Some BOINC projects are based at universities and research labs while others are independent areas of research or interest.^[23]