This article was accepted into the corpus but its outbound wikilinks were never NER-processed — typical at the deepest BFS hop or when the run's entity cap was reached. No expansion funnel to show.
| IBM 370 | |
|---|---|
| Name | IBM System/370 |
| Developer | International Business Machines |
| Introduced | 1970 |
| Discontinued | 1990s |
| Architecture | CISC, 32-bit mainframe |
| Successors | IBM System/390, IBM zSeries |
IBM 370
The IBM System/370 family was a series of mainframe computers introduced by International Business Machines in 1970 as a successor to an earlier mainframe line. It played a central role in enterprise computing through the 1970s and 1980s, influencing banking, telecommunications, aerospace, and government data centers. Major vendors, corporate customers, and research institutions adopted the platform for transaction processing, scientific computing, and time-sharing applications.
Development of the System/370 lineage followed earlier initiatives at International Business Machines and drew on design work associated with earlier mainframes and research projects. Engineering decisions reflected lessons from projects at companies and institutions such as Honeywell, Control Data Corporation, General Electric, Bell Labs, and Digital Equipment Corporation. The release occurred amid competition with systems like the UNIVAC 1108, CDC 6600, DEC PDP-11, and systems promoted by Fujitsu and Burroughs Corporation. Design priorities included backward compatibility for customers running software on IBM System/360, improved reliability influenced by practices from AT&T operations, and enhanced input/output subsystems inspired by work at NASA and MIT. Management and marketing were coordinated across corporate units influenced by executives who had worked with Thomas J. Watson Jr. and were responding to regulatory contexts shaped by agencies such as the Federal Communications Commission and economic pressures from trade partners like Japan.
The hardware architecture emphasized a 32-bit general register model with byte-addressable memory and an instruction set architecture supporting binary, decimal, and floating-point arithmetic. Architectural elements reflected research parallels with efforts at Stanford University, Massachusetts Institute of Technology, and University of California, Berkeley. The instruction set included operations for data movement, arithmetic, logical operations, and control flow, enabling compilers and tools developed at companies such as Microsoft-era collaborators and research groups at Carnegie Mellon University. Microarchitectural innovations paralleled contemporaneous work at Intel and Motorola on bus designs and cache hierarchies, while channel architecture and direct memory access resembled I/O models used by Honeywell and DEC. The System/370 introduced features later enhanced by concepts from IBM Research laboratories, influenced by academic publications from ACM and IEEE conferences.
The family encompassed a range of models from lower-end systems aimed at regional data centers to high-end configurations for large institutions. Vendors and customers configured machines with varying memory sizes, processor cycle times, and peripheral options marketed similarly to product lines from Fujitsu, Hitachi, Amdahl Corporation, and Unisys. High-performance variants competed with supercomputer efforts like those at Cray Research and were deployed in scientific centers associated with Argonne National Laboratory, Lawrence Livermore National Laboratory, and university supercomputing centers. Third-party manufacturers produced compatible systems and add-ons resembling peripheral ecosystems offered by StorageTek and Emulex.
A broad software ecosystem developed around the hardware, including operating systems, database management systems, transaction monitors, and development tools. Major operating systems ran on the line and were developed or supported by organizations such as IBM itself, independent software vendors, and academic groups; these included batch and time-sharing systems informed by research at Project MAC and service bureaus operated by firms like Electronic Data Systems. Database and middleware vendors serving the platform had business links with Oracle Corporation, Cullinet, and early SAS Institute deployments in analytics centers. Programming languages and toolchains from bodies like X/Open and compiler research at Princeton University helped shape FORTRAN, COBOL, PL/I, and assembler toolsets used widely by banks, insurance companies, and governments including those of United Kingdom and United States agencies.
Performance scaled with semiconductor advances and system architecture enhancements influenced by developments at Intel and Texas Instruments. The line supported mission-critical workloads for major corporations such as Bank of America, American Express, AT&T, General Motors, and governmental systems involving Department of Defense contracting and census operations. Transaction-processing capabilities underpinned credit card networks and airline reservation systems tied to firms like SABRE and logistics operations with partners such as Federal Express. The platform’s reliability and serviceability informed standards and procurement decisions at multinational corporations and central banks in countries such as Germany and Japan.
The architectural and operational concepts pioneered in the family influenced subsequent IBM platforms like successors designed for enterprise and cloud computing, and informed industry standards developed by groups such as ANSI and ISO. Concepts from the platform echo in modern mainframe-class systems at IBM z Systems, virtualization approaches seen in hypervisor work at VMware and project groups at Xerox PARC, and transaction processing ideas used by payment networks and large-scale database providers like SAP and Oracle Corporation. Academic curricula at institutions such as MIT, Stanford University, and Carnegie Mellon University continue to teach principles traceable to the platform’s design, and historic installations appear in museums including exhibits curated by the Computer History Museum and institutions preserving computing heritage.