IBM 700/7000 series

From Wikipedia, the free encyclopedia
  (Redirected from FORTRAN Assembly Program)
Jump to: navigation, search
An IBM 704 mainframe at NACA in 1957

The IBM 700/7000 series was a series of large-scale (mainframe) computer systems made by IBM through the 1950s and early 1960s. The series included several different, incompatible processor architectures. The 700s used vacuum tube logic and were made obsolete by the introduction of the transistorized 7000s. The 7000s, in turn, were eventually replaced by System/360, which was announced in 1964. However the 360/65, the first 360 powerful enough to replace 7000s, did not become available until November 1965. Early problems with OS/360 and the high cost of converting software kept many 7000s in service for years afterward.

Architectures[edit]

The IBM 700/7000 series had six completely different ways of storing data and instructions:

The 700 class used vacuum tubes, the 7000 class was transistorized. All machines (like most other computers of the time) used magnetic core memory; except for early 701 and 702 models, which used CRT memory.

Software compatibility issues[edit]

Early computers were sold without software. As operating systems began to emerge, having four different mainframe architectures plus the 1400 midline architectures became a major problem for IBM since it meant at least four different programming efforts were required.

The System/360 combined the best features of the 7000 and 1400 series architectures into a single design. However, some 360 models had optional features that allowed them to emulate the 1400 and 7000 instruction sets in microcode. One of the selling points of the System/370, introduced in mid-1970, was improved 1400/7000 series emulation, which could be done under operating system control rather than shutting down and restarting in emulation mode as was required on the 360s.

Peripherals[edit]

While the architectures differed, the machines in the same class shared electronics technologies and generally used the same peripherals. Tape drives were 7-track format, with the IBM 727 used on vacuum tube machines and the 729 on transistor machines. Both the vacuum tube and most transistor models used the same card readers, card punches and line printers that were introduced with the 701. These units, the IBM 711, 721 and 716, were based on IBM accounting machine technology and even included plugboard control panels. They were relatively slow and it became common for 7000 series installations to include an IBM 1401, with its much faster peripherals, to do card-to-tape and tape-to-line-printer operations off-line. Three later machines, the 7010, the 7040 and the 7044, adopted peripherals from the midline IBM 1400 series. Some of the technology for the 7030 was used in data channels and peripheral devices on other 7000 series computers, e.g., 7340 Hypertape.

First architecture (701)[edit]

Main article: IBM 701
IBM 701 operator's console
An IBM 704 installation

Known as the Defense Calculator while in development in the IBM Poughkeepsie Laboratory, this machine was formally unveiled April 7, 1953 as the IBM 701 Electronic Data Processing Machine.

Data formats[edit]

Numbers were either 36 bits or 18 bits long, only fixed point. (See: Why 36 bits?)

Instruction format[edit]

Instructions were 18 bits long, single address.

  • Sign (1 bit) – Whole word (-) or Half word (+) operand address
  • Opcode (5 bits) – 32 instructions
  • Address (12 bits) – 4096 Half word addresses

To expand the memory from 2048 to 4096 words, a 33rd instruction was added that used the most significant bit of its address field to select the bank. (This instruction was probably created using the "No OP" instruction, which appears to have been the only instruction with unused bits, as it originally ignored its address field. However documentation on this new instruction is not currently available.)

Registers[edit]

Processor registers consisted of:

  • AC  - 38-bit Accumulator
  • MQ – 36-bit Multiplier-Quotient

Memory[edit]

2,048 or 4,096 – 36-bit binary words with six-bit characters

Scientific Architecture (704/709/7090/7094)[edit]

IBM 7090's at NASA's Project Mercury, 1962.

IBM's 36-bit scientific architecture was used for a variety of computation-intensive applications. First machines were the vacuum-tube 704 and 709, followed by the transistorized 7090, 7094, 7094-II, and the lower cost 7040 and 7044. The ultimate model was the Direct Coupled System (DCS) consisting of a 7094 linked to a 7044 that handled input and output operations.

Data formats[edit]

Numbers were 36 bits long, both fixed point and floating point. (See: Why 36 bits?)

  • Fixed point numbers were stored in binary sign/magnitude format.
  • Single precision floating point numbers had a magnitude sign, an 8-bit excess-128 exponent and a 29 bit magnitude
  • Double precision floating point numbers, introduced on the 7094, had a magnitude sign, a 17-bit excess-65536 exponent, and a 54 bit magnitude
  • Alphameric characters were 6-bit BCD, packed six to a word.

Instruction format[edit]

The basic instruction format was a 3-bit prefix, 15-bit decrement, 3-bit tag, and 15-bit address. The prefix field specified the class of instruction. The decrement field often contained an immediate operand to modify the results of the operation, or was used to further define the instruction type. The three bits of the tag specified three (seven in the 7094) index registers, the contents of which were subtracted from the address to produce an effective address. The address field either contained an address or an immediate operand.

Registers[edit]

Vacuum tube logic module from a 700 series IBM computer.
Transistorized IBM Standard Modular System (SMS) card used in the 7000 series.

Processor registers consisted of:

  • AC  - 38-bit Accumulator
  • MQ – 36-bit Multiplier-Quotient
  • XR  - 15-bit Index Registers (three or seven)
  • SI    - 36-bit Sense Indicator

The accumulator (and multiplier-quotient) registers operated in signed magnitude format.

The Index registers operated using two's complement format and when used to modify an instruction address were subtracted from the address in the instruction. On machines with three index registers, if the tag had 2 or 3 bits set (i.e. selected multiple registers) then their values were ORed together before being subtracted. The IBM 7094, with seven index registers had a "compatibility" mode to permit programs from earlier machines that used this trick to continue to be used.

The Sense Indicators permitted interaction with the operator via panel switches and lights.

Memory[edit]

  • 704
    • 4,096 or 8,192 or 32,768 – 36-bit binary words with six-bit characters
  • 709, 7090, 7094, 7094 II, 7040, 7044
    • 32,768 – 36-bit binary words with six-bit characters

Input/output[edit]

The 709/7090 series used Data Synchronizer Channels for high speed input/output, such as tape and disk. The basic DSCs, e.g., 7607, executed their own simple programs from the computer memory that controlled the transfer of data between memory and the I/O devices; the more advanced 7909 supported more sophisticated channel programs and an 8-bit interface. Punched card I/O and high speed printing were often performed by transferring magnetic tapes to an off-line IBM 1401. Later, the data channels were used to connect a 7090 to a 7040 or a 7094 to a 7044 to form the IBM 7094/7044 Direct Coupled System (DCS). In that configuration, the 7044, which could use faster 1400 series peripherals, primarily handled I/O.

FORTRAN assembly program[edit]

The FORTRAN Assembly Program (FAP) was an assembler for the 709, 7090, and 7094 under IBM's makeshift FMS (Fortran Monitor System) operating system. An earlier assembler was SCAT (SHARE Compiler-Assembler-Translator). Macros were added to FAP by Bell Laboratories (BE-FAP), and the final 7090/7094 assembler was IBMAP, under IBSYS.

Its pseudo-operation BSS, used to reserve memory, is the origin of the common name of the "BSS section", still used in many assembly languages today for designating reserved memory address ranges of the type not having to be saved in the executable image.

Commercial architecture (702/705/7080)[edit]

IBM 702
IBM 7080

The IBM 702 and IBM 705 were similar and the 705 could run many 702 programs without modification, but they were not completely compatible.

The IBM 7080 was a transistorized version of the 705, with various improvements. For backward compatibility it could be run in 705 I mode, 705 II mode, 705 III mode, or full 7080 mode.

Data format[edit]

Data was represented by a variable-length string of characters terminated by a Record mark.

Instruction format[edit]

Five characters: one character opcode and four character address – OAAAA

Registers[edit]

  • 702
    • two Accumulators (A & B) – 512 characters
  • 705
    • one Accumulator – 256 characters
    • 14 auxiliary storage units – 16 characters
    • one auxiliary storage unit – 32 characters
  • 7080
    • one Accumulator – 256 characters
    • 30 auxiliary storage units – 512 characters
    • 32 communication storage units – 8 characters

Memory[edit]

  • 702
    • 2,000 to 10,000 characters in Williams tubes (in increments of 2,000 characters)
    • Character cycle rate – 23 microseconds
  • 705 (models I, II, or III)
    • 20,000 or 40,000 or 80,000 characters of Core memory
    • Character cycle rate – 17 microseconds or 9.8 microseconds
  • 7080
    • 80,000 or 160,000 characters of Core memory
    • Character cycle rate – 2.18 microseconds

1400 series architecture (7010)[edit]

The 700/7000 commercial architecture inspired the very successful IBM 1400 series of mid-sized business computers. In turn, IBM later introduced a mainframe version of the IBM 1410 called the IBM 7010.

Data format[edit]

Data was represented by a variable length string of characters terminated by a Wordmark.

Instruction format[edit]

Variable length: 1, 2, 6, 7, 11, or 12 characters.

Registers[edit]

None, all instructions operated on memory. However, fifteen five-character fields in fixed locations in low memory could be treated as index registers, whose values could be added to the address specified in an instruction. Also, certain internal registers that would today be invisible, such as the addresses of the characters being currently processed, were exposed to the programmer.

Memory[edit]

100,000 characters.[1]

Decimal architecture (7070/7072/7074)[edit]

IBM 7074

The IBM 7070, IBM 7072, and IBM 7074 were decimal, fixed-word-length machines. They used a ten digit word like the smaller and older IBM 650, but were not instruction set compatible with the 650.

Data format[edit]

  • Word length – 10 decimal digit plus sign
  • Digit encoding – two-out-of-five code
  • Floating point – optional. Two digit exponent.
  • Three signs for each word – Plus, Minus and Alpha
    • Plus and Minus indicated 10 digit numeric values.
    • Alpha indicated 5 characters of text coded by pairs of digits. 61 = A, 91 = 1.

Instruction format[edit]

  • All instructions one word
  • 2 digit op code (including sign, Plus or Minus only)
  • 2 digit index register
  • 2 digit field control – allows selecting sets of digits, shifting left or right
  • 4 digit address

Registers[edit]

  • All registers one word, could also be addressed as memory
  • Accumulators – 3 (addresses 9991, 9992, and 9993 – standard; 99991, 99992, and 99993 – extended 7074)
  • Program register – 1 (address 9995 – standard; 99995 – extended 7074)
    • Addressable from console only. Stores current instruction.
  • Instruction counter – 1 (address 9999 – standard; 99999 – extended 7074)
    • Addressable from console only.
  • Index registers – 99 (addresses 0001-0099)

Memory[edit]

  • 5000 to 9990 words (standard)
  • 15000 to 30000 words (extended 7074)
  • Access time – 6 microseconds (7070/7072), 4 microseconds (7074)
  • Add time – 72 microseconds (7070), 12 microseconds (7072), 10 microseconds (7074)

Timeline[edit]

year category logic memory
decimal commercial scientific supercomputer
1952 IBM 701 vacuum tubes Williams tubes
1953 IBM 702
1954 IBM 705 IBM 704 core memory
1958 IBM 7070 IBM 709
1959 IBM 7090 transistors
1960 IBM 7074
1961 IBM 7080 IBM 7030
1962 IBM 7010 IBM 7094
1963 IBM 7040
IBM 7044
1964 IBM 7094 II

An IBM 7074 was used by the U.S. Internal Revenue Service in 1962.[1]

The IBM 7700 Data Acquisition System was not a member of the IBM 7000 series, despite its number and its announcement date of December 2, 1963.

Performance[edit]

All of the 700 and 7000 series machines predate standard performance measurement tools such as the Whetstone (1972), Dhrystone (1984), LINPACK (1979), or Livermore loops (1986) benchmarks.

In the table below, the Gibson and Knight measurements report speed, where higher numbers are better; the TRIDIA measurement reports time, where lower numbers are better.

Model Gibson
mix
KIPS
Knight
Index
scientific[2]
TRIDIA program
(FORTRAN)
(seconds)[3]
IBM 705 m1,2 0.50
IBM 705 m3 0.38
IBM 709 21
IBM 7030 372 15.58
IBM 7040 148
IBM 7044 109 74
IBM 7090 139 66
IBM 7094 176 31.35
IBM 7094 II 257 217 16.50

References[edit]

  1. ^ "Big-Brother 7074 is watching you" by Robert Gannon, Popular Science, March 1963
  2. ^ M. Phister, Jr., Data Processing Technology and Economics, 2nd ed., 1979; Table II.2.11.1
  3. ^ "Final Report on 64/6600 FORTRAN Version 3.0", Control Data Corporation, 1966 June 6; section I.B, pp. 3-4; at bitsavers.org

External links[edit]