← Back to Projects
⚙️

BOINC

Platform Active Platform / Infrastructure CPU + GPU Since 2002
https://boinc.berkeley.edu ↗

BOINC (Berkeley Open Infrastructure for Network Computing) is the world's leading open-source platform for volunteer and grid computing. Developed at the UC Berkeley Space Sciences Laboratory by David Anderson — who previously helped create SETI@home — BOINC provides the complete middleware stack that connects volunteers' idle computers to scientific research projects around the world.

The platform handles every aspect of distributed computation: packaging scientific applications for multiple operating systems, distributing work units to volunteers, scheduling tasks based on each computer's capabilities and the user's preferences, collecting and validating results through redundant computation, and awarding credits to participants. BOINC supports CPU, GPU (NVIDIA and AMD), and multi-core workloads across Windows, macOS, Linux, and Android.

BOINC's architecture is designed around a client-server model. Each research project runs its own BOINC server, which generates work units and distributes them through the BOINC client installed on volunteers' machines. A single volunteer can attach to multiple projects simultaneously, with BOINC managing the allocation of computing time according to user-defined preferences. This multi-project capability is one of BOINC's key advantages over earlier single-project systems.

Over 30 active research projects currently use BOINC, spanning astrophysics, molecular biology, climate science, mathematics, and more. At its combined peak, BOINC volunteers have provided computing power rivaling the world's most powerful supercomputers. The platform has been cited in well over a thousand scientific publications and has enabled research that would be financially or logistically impossible using conventional computing resources.

BOINC is free and open-source (LGPL licensed), making it accessible to any research institution that needs large-scale computing but lacks the budget for dedicated supercomputer time.