CDC 6000 series

CDC 6000 series
Name	CDC 6000 series
Developer	Control Data Corporation
Family	CDC mainframe
Released	1964
Discontinued	1975
Cpu	6600, 6400, 6500
Memory	core memory
Os	SCOPE, KRONOS, COS
Weight	varies
Predecessor	CDC 3000 series
Successor	CDC 7600

CDC 6000 series The CDC 6000 series was a line of high-performance mainframe computers produced by Control Data Corporation in the 1960s, notable for pioneering supercomputing concepts and influencing contemporary systems architecture. Designed under the leadership of Seymour Cray and marketed by William Norris, the series competed with offerings from International Business Machines, UNIVAC, and Honeywell while serving scientific, military, and industrial customers including NASA, Los Alamos National Laboratory, and U.S. Department of Defense. Its engineering impacted later machines at Cray Research and informed projects at institutions like Stanford University, MIT, and Argonne National Laboratory.

Overview

The 6000 series introduced a multiprocessor layout featuring a high-speed central processor and multiple peripheral processors, reflecting design goals championed by Seymour Cray and implemented at Control Data Corporation facilities in Minneapolis, Minnesota and Plymouth, Minnesota. It entered commercial service amid competition from IBM System/360, UNIVAC 1108, and Honeywell 6000 series models, winning contracts from organizations such as Lawrence Livermore National Laboratory, General Electric, Bell Labs, and RCA. Key figures associated with its development include James Thornton and Gordon Bell, both of whom later influenced computing at DEC and Microsoft Research.

Architecture and Design

The architecture separated a high-speed central processor from ten peripheral processors, a scheme inspired by concepts from John von Neumann-era designs and contemporary work at Bell Labs and MIT Lincoln Laboratory. The central processor employed a 60-bit word size and a 10,000 to 12,000 cycles-per-second clock in early models, while peripheral processors handled I/O under control programs similar to ideas from Fernando Corbató’s work at MIT and the Compatible Time-Sharing System. The design used magnetic core memory and modular cabinets reminiscent of installations at Argonne National Laboratory and Brookhaven National Laboratory. Engineering tradeoffs paralleled research at Los Alamos National Laboratory and influenced later vector and pipeline approaches used by Cray Research and IBM Research.

Models and Variants

Principal members included the 6600 flagship designed by Seymour Cray, the 6400 targeted for midrange workloads, and the 6500 series aimed at scaled configurations for installations at NASA Ames Research Center and Sandia National Laboratories. Variants and upgrades were installed at Oak Ridge National Laboratory, Fermilab, and corporate sites such as Ford Motor Company and General Motors. Successor and related lines—developed by Control Data Corporation engineers who would later join Cray Research and CDC Cyber teams—connected the series to machines like the CDC 7600, CDC Cyber 70, and research efforts at University of California, Berkeley and Carnegie Mellon University.

Operating Systems and Software

The series ran operating systems developed at Control Data Corporation including SCOPE, KRONOS, and COS, whose designs were informed by concepts from Multics work at Bell Labs and MIT. Compilers and tools included FORTRAN implementations influenced by standards from IBM, debugging systems inspired by practices at DEC and Bell Labs, and scientific libraries used by researchers at Lawrence Livermore National Laboratory and Los Alamos National Laboratory. The software ecosystem interfaced with batch processing centers at Sandia National Laboratories, timesharing experiments at Stanford University, and visualization efforts at Argonne National Laboratory.

Performance and Benchmarks

At introduction the flagship model outperformed contemporaries such as IBM 7090, UNIVAC 1108, and CDC 3000 series installations on floating-point workloads, reflecting innovations that paralleled research at Los Alamos National Laboratory and benchmarking practices developed at National Institute of Standards and Technology. Benchmarks emphasized scientific computing metrics derived from workloads at NASA, Fermilab, and Lawrence Livermore National Laboratory and compared integer and floating-point throughput against machines from IBM Research and Bell Labs. The architecture’s effective instruction-level parallelism presaged later pipeline and vector benchmarks used by Cray Research and academic groups at MIT and University of Illinois Urbana-Champaign.

Historical Impact and Legacy

The 6000 series shaped the trajectory of supercomputing, directly influencing the establishment of Cray Research and the designs of later machines such as the CDC 7600 and systems built by Cray Inc.. Its operational presence at national labs including Los Alamos National Laboratory, Lawrence Livermore National Laboratory, Oak Ridge National Laboratory, and Sandia National Laboratories affected computational science in fields tied to initiatives at NASA, Fermilab, CERN, and industrial research at General Electric and Bell Labs. Personnel associated with the series—among them Seymour Cray, Gordon Bell, and William Norris—went on to influence computing at DEC, IBM, Microsoft Research, and academic institutions such as Stanford University and Carnegie Mellon University, while archival machines and emulators persist in museums like the Computer History Museum and collections at Smithsonian Institution.

Category:Mainframe computers