Generated by GPT-5-mini| Multics | |
|---|---|
| Name | Multics |
| Developer | Massachusetts Institute of Technology, General Electric, Honeywell |
| Released | 1969 |
| Discontinued | 1988 |
| Latest release | Honeywell Multics releases (various) |
| Programming language | PL/I, Assembly language |
| Platform | GE-600 series, Honeywell 6000 series, Honeywell DPS-8 |
| License | Proprietary |
Multics Multics was a pioneering time-sharing operating system developed to explore large-scale interactive computing, virtualization, and secure multi-user access. Its design influenced later systems and involved collaborations among Massachusetts Institute of Technology, Bell Labs, and General Electric, impacting research at MIT Project MAC, commercial strategies at Honeywell, and later academic work at Princeton University and Stanford University.
Multics aimed to provide a secure, reliable, and highly available environment for multiple users across Project MAC, MIT, and Honeywell installations, emphasizing goals from Fernando J. Corbató-led time-sharing research and ideas propagated through J. Presper Eckert-era debates about interactive computing. Designers targeted persistent single-level storage inspired by concepts emerging at Cambridge University and RAND Corporation discussions, seeking to supplant batch-oriented systems used by IBM and DEC with interactive services promoted by ARPA and Advanced Research Projects Agency sponsors.
Multics featured a hierarchical file system, dynamic linking, and ring-based protection that drew on theoretical work from Gerald J. Popek and Robert M. Tomasulo influences and ideas explored at Stanford Research Institute. Its architecture supported segmented memory management, demand paging, and symmetric multiprocessing compatible with GE-645 hardware, integrating virtual memory concepts similar to those investigated at University of California, Berkeley and University of Cambridge Computer Laboratory. The system implemented modular kernels, a shell environment that influenced later shells at Bell Labs and GNU Project contributors, and a security model paralleling formal models discussed at National Security Agency and RAND Corporation workshops.
Development began under a contract linking MIT Project MAC, Bell Labs, and General Electric in the mid-1960s, with milestones including early prototypes on GE-600 series hardware, production releases during the late 1960s, and commercial support by Honeywell after acquisitions involving Honeywell and General Electric. Key figures included Fernando J. Corbató, M. Douglas McIlroy, and engineers who later joined projects at Bell Labs and Xerox PARC. Significant events include the 1965 design meetings at Cambridge, Massachusetts, procurement negotiations with USAF and National Science Foundation, and declines in Bell participation leading to the formation of successor projects in the early 1970s.
Multics directly influenced the design of Unix at Bell Labs, inspired language and OS research at MIT Artificial Intelligence Laboratory, and informed commercial operating systems developed by DEC, IBM, and Honeywell. Concepts from Multics shaped security research at National Computer Security Center, file system work at CMU (Carnegie Mellon University), and virtualization advances at Stanford University and University of Cambridge. Alumni of Multics contributed to projects at Sun Microsystems, Microsoft, and Google, and ideas from Multics are cited in standards work at IEEE and ISO committees.
Multics ran on GE-645 and subsequent Honeywell 6000 series machines and was deployed at research centers including MIT, DARPA-funded facilities, Bell Telephone Laboratories, Honeywell corporate sites, and government installations such as NSA-affiliated operations and US Air Force labs. Academic implementations influenced teaching at Harvard University, Princeton University, and University of California, Berkeley, while commercial installations serviced users at Honeywell customer sites and corporate research groups in Silicon Valley and Cambridge, Massachusetts.
Multics pioneered ring-based access control, segmented memory protections, and live upgrade techniques that foreshadowed formal security models debated at National Security Agency and analyzed in literature by David P. Reed and Roger R. Schell. Its single-level store and auditing facilities influenced evaluation criteria used by Rainbow Books-era certification frameworks and inspired operational security practices at NSA and within Defense Advanced Research Projects Agency-sponsored security projects. Reliability features such as process isolation, fault containment, and online system maintenance were precursors to high-availability practices adopted by AT&T and IBM in telecommunications and enterprise systems.
Contemporaries and later commentators from Bell Labs, DEC, and IBM praised Multics for its ambition and depth while criticizing its complexity, performance on contemporary GE-600 series hardware, and commercial viability compared to simpler systems like Unix and TOPS-10. Economists and procurement officials at USAF and National Science Foundation questioned its cost-effectiveness, while software engineers at Xerox PARC and Stanford University debated the trade-offs between feature-rich monolithic designs and minimalist approaches championed by Ken Thompson and Dennis Ritchie allies.
Category:Operating systems