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.
| CDC STAR-100 | |
|---|---|
| Name | CDC STAR-100 |
| Developer | Control Data Corporation |
| Release | 1975 |
| Type | Vector supercomputer |
| Cpu | Scalar and vector pipelines |
| Speed | Up to 100 MFLOPS (peak) |
| Memory | Up to 8 MB core/semiconductor memory |
| Predecessor | CDC 6000 series |
| Successor | CDC Cyber 205 |
CDC STAR-100 The CDC STAR-100 was a vector supercomputer produced by Control Data Corporation in the mid-1970s. Designed to compete with contemporaries such as Cray Research, SSE (supercomputer), and architectures influenced by work at Los Alamos National Laboratory and Lawrence Livermore National Laboratory, the STAR-100 emphasized pipeline vector processing for scientific workloads. Its architecture and commercial fate intersected with developments at IBM, Hewlett-Packard, Sandia National Laboratories, and academic centers like MIT, Stanford University, and University of Illinois Urbana-Champaign.
The STAR-100 employed a long-vector pipeline influenced by earlier designs at Control Data Corporation and research from Argonne National Laboratory, featuring vector registers and a memory subsystem comparable to systems from Cray Research and Burroughs Corporation. Its instruction set and microarchitecture were informed by needs identified at NASA, Oak Ridge National Laboratory, and CERN. The machine supported high-throughput floating-point operations comparable in aspiration to the CDC 6600 family and used semiconductor logic similar to modules in Digital Equipment Corporation systems. Memory bandwidth and latency trade-offs mirrored concerns in projects at Bell Labs, Sandia National Laboratories, and National Center for Supercomputing Applications.
Development was led by engineering teams at Control Data Corporation headquarters in Minneapolis, drawing on talent with experience from projects at Los Alamos National Laboratory, Argonne National Laboratory, and collaborations with University of Minnesota. Manufacturing processes used techniques prevalent in the 1970s semiconductor and module assembly industries, comparable to practices at Intel, Texas Instruments, and Motorola. Corporate strategy debates involved executives with ties to Sperry Corporation and General Electric, while sales and procurement engaged agencies like Department of Energy laboratories, research centers including Caltech, and aerospace contractors such as Lockheed Martin and Boeing.
CDC offered evolutions and field upgrades that attempted to address performance gaps relative to machines from Cray Research and Fujitsu. Later configurations incorporated larger memory and faster scalar units analogous to enhancements seen in CDC Cyber models and in efforts at IBM Research and Hitachi to boost vector throughput. Customer-driven adaptations paralleled upgrade programs at Los Alamos National Laboratory, Lawrence Livermore National Laboratory, and university centers including University of Cambridge and Imperial College London.
The STAR-100 was deployed at national laboratories, universities, and industrial research sites, joining computing centers alongside systems from Cray Research, IBM, Fujitsu, and Burroughs Corporation. Installations appeared at Sandia National Laboratories, Lawrence Livermore National Laboratory, Argonne National Laboratory, and universities such as MIT, Stanford University, and University of California, Berkeley. Operational experience influenced procurement decisions at National Science Foundation-funded centers and informed comparative studies conducted at NASA Ames Research Center and European Space Agency facilities.
In practice the STAR-100’s vector pipelines delivered high peak floating-point rates comparable in stated metrics to offerings from Cray Research and Fujitsu but faced real-world limitations due to memory latency and compiler support, issues also observed in contemporaneous systems at IBM and Burroughs Corporation. Performance tuning and benchmarking involved suites and practices used at Los Alamos National Laboratory, Argonne National Laboratory, NASA, and academic centers such as Stanford University and University of Illinois Urbana-Champaign. Compiler technology from teams with experience at Bell Labs and IBM Research was critical to extracting throughput, while applications in computational fluid dynamics, weather modeling, and nuclear simulation paralleled work at National Center for Atmospheric Research, NOAA, and Lawrence Livermore National Laboratory.
The STAR-100 influenced subsequent vector and parallel designs, informing engineering choices at Control Data Corporation and competitors like Cray Research and IBM. Lessons learned affected procurement policies at Department of Energy laboratories, the evolution of scientific computing curricula at MIT and Stanford University, and the design of later machines at Cray Research, Fujitsu, and Hitachi. Its commercial record contributed to industry analyses by firms such as Gartner and academic studies published in venues associated with ACM and IEEE. The machine remains cited in retrospectives on supercomputing history alongside milestones like the CDC 6600, Cray-1, and the rise of parallel architectures at Los Alamos National Laboratory and Oak Ridge National Laboratory.