International Business Machines Corporation (IBM)
||1 1 910 0
||On BYU asset tag on front left of object
overall: 67 in x 64 in x 29 1/2 in
Gwen Bell tracked the provenance on this machine in early 1999. It is S/N 1. It was initially at Los Alamos National Laboratory. Len Shustek writes: IBM Stretch facts The IBM 7030, or Stretch, was the result of an intensive R&D project started at IBM in 1955. The goal: build a "super-computer" 100 to 200 times as powerful as anything yet built. The premier customer: Los Alamos Scientific Laboratory (run by the Atomic Energy Commission), which was designing A-bombs and H-bombs. It took six years, and was delivered to Los Alamos in 1961. It "stretched" the state-of-the-art on many fronts: the first major IBM product to use transistors instead of vacuum tubes (show tube and transistor?), big fast disk drives with multiple read/write arms (32MB), and big high-speed magnetic core memory (2 MB). But especially architectural innovations: pipelining (allowing multiple computer instructions to execute at once in different parts of the machine), lookahead (fetching and decoding up to six instructions before they are needed, to get a jump-start.), and multiprogramming (running more than one application at a time) and error-correction (having the machine diagnose and correct its own faults.). It was the most complex electronic circuitry ever designed, and in fact the first to use an earlier computer (the IBM 704) to help in the design. But it failed its primary goal of being 200 or even 100 times faster: only about 25- 50 times. So IBM couldn't charge the $13M they originally wanted, but only $8M, and at that price they would lose money on every one. Only seven other Stretch machines were built after the one that went to Los Alamos, all to government agencies (like the Weather Service for charting the path of storms) or government contractors (like MITRE). One specially-modified machine called "Harvest" was sold to the National Security Agency for code-breaking work. Stretch was a commercial failure, and president TJ Watson was embarrassed and furious. Steve "Red" Dunwell, the project's head, was exiled to the research lab, never to work on the design of computers again. But eventually he was was made an IBM Fellow, and received from Watson as close to an apology for the company's treatment of his reputation as Watson was capable of giving, becauseî Many of the innovations in Stretch were used in later IBM computers that were great successes (7090, 360). They also served as inspiration for many non-IBM computers, and many of the Stretch-like features disappeared for a while before appearing in other big computers. They have only recently been incoporated in modern microprocessors, which are now finally big enough to do on a signle chip what Stretch did in 1960. Label copy is as follows: Number Produced: 9 Price: $6-8 Million Project Start: 1954 Project Leaders: S.W. Dunwell; Gene Amdahl, John Backus, Werner Buchholz, B.O. Evans, Jerrier Haddad, Lloyd Hunter, Ralph Palmer, and John Sheldon First Delivery: April 1961 to the Los Alamos Scientific Laboratory Software: Algebraic and Fortran compiler Use: Large scale scientific research, for example: nuclear reactor design, hydrodynamic problems, problems in nuclear physics. Achievements: Techniques for parallel processing and multi-programming were interleaved memories, instruction look-up units, overlapping fetch and execute instructions, interrupt handling and address monitoring. The 7030 also introduced an 8-bit byte for character representation, up to 256 characters could be represented. The magnetic core memory developed for the STRETCH was also used on the IBM 7090.
I/O/console / panel
Los Alamos National Laboratory; IBM Project Stretch; IBM 7030 (Computer)