Method of representing and processing data in a base-b numbering system, computer and computer network

Abstract

Proposed is a method in which all information transmitted over a system bus to and from a processor is represented in the form of multi-bit numbers, with each bit being represented in a binary positional numbering system, wherein each number in the system bus is represented in the form of a given n-bit number in a base-b positional numbering system; in each bit, each digit having one of the values in a range of from 0 to b -1 is represented by a k-bit number in a binary positional numbering system; each bit in this number is represented by an optical laser beam; each optical laser beam travels along a separate optically isolated optical fibre; together, all of the optically isolated optical fibres in each bit form a bit bundle; together, all of the bit bundles form a system bus. Also proposed are a computer and a computer network which operate on the above method. The advantage of the inventions is a decrease in the number of bits in data, address and command buses and thus a reduction in the number of data processing operations and therefore also the machine time as a result of an increase in the information content of each bit.

Description

[0001] A method of representing and processing information in the b-ary number system, computer and information and computing network.

[0002] DESCRIPTION

[0003] [1] FIELD OF TECHNOLOGY

[0004] [2] The invention relates to computing technology.

[0005] [3] LEVEL OF TECHNOLOGY

[0006] [4] In modern computers, computer networks and systems, all initial information (numeric, text, graphic, etc.) is transformed and ultimately presented in the digital format of the positional number system (hereinafter PNS), when the same numerical sign (digit) in the number record has different values ​​depending on the place (digit) where it is located. Binary, ternary, decimal, b-ary PNS are known. The main feature of the PNS is the presence of zero in each digit, and the base b can be any natural number.

[0007] [5] The method described in (1) is known, where in computers the digital format is presented in binary form (code), when each binary digit takes one of two possible values ​​- logical "0" or "1". This is due to the fact that it is most easily implemented by technical devices that have two quasi-stable (metastable) states (on / off, connected / disconnected, transistor open / closed).

[0008] The main disadvantage of these known solutions is the minimal information content of each digit.

[0009] [6] The closest to the proposed method is the method described in patent BY 6137 "Optoelectronic adder" (2), which uses two optically coupled pairs of LEDs and a laser. The disadvantage of the method is the small number of output signal levels, they do not increase the information content of either each digit in the numbers or the numbers themselves, since they are expressed in the binary number system. 231123-01-07 info

[0010] [7] Modern computers are based on the classical architecture proposed in 1946 by the American mathematician John von Neumann (3), which describes the principle in which information processing in a computer is carried out in a binary PSS. The closest in functional capabilities to the proposed computer invention is a computer based on ternary logic, developed by N.P. Brusentsov. The described computer (4) does not have a sufficiently high information content of each digit.

[0011] [8] For information and computing networks there are tasks that are difficult or impossible to solve without organizing information communications between different computers: transferring information over long distances; sharing expensive hardware, software or information resources among several computers.

[0012] [9] Computer networks (5) are known, representing a set of computers and other devices connected by communication lines and exchanging information with each other, represented in binary PSS. The disadvantage is the presence of various protocols and insufficient speed of information exchange between individual computers due to the use of only binary PSS.

[0013]

[0010] The closest known technical solution to the proposed solution for information and computing networks is the network architecture for high-speed data transmission over fiber-optic lines ATM (Asynchronous Transfer Mode), described in (6).

[0014] The disadvantage of this network is its low information content due to the use of only binary PSS.

[0015]

[0011] SUMMARY OF THE INVENTION

[0016]

[0012] The purpose of the proposed inventions is to reduce the number of bits in the buses (data, address and commands) and thereby reduce the number of information processing operations, and therefore machine time, by increasing the information content of each bit.

[0017]

[0013] This goal is achieved by proposing a method, a computer and an information and computing network that uses a method for representing and processing data in a computer, where:

[0018]

[0014] - all information sent via the system bus to and from the processor is represented in the form of multi-bit numbers, with each bit represented in a positional number system:

[0019]

[0015] - each number in the system bus is represented as a conditional n-bit number in the b-ary positional number system,

[0016] - each digit in each digit, having one of the values ​​in the range from 0 to b-1, is represented by a K-bit number in the binary positional number system,

[0020]

[0017] - each digit in this number is represented by an optical laser beam,

[0021]

[0018] - each optical laser beam travels along a separate optically isolated glass fiber,

[0022]

[0019] - all together, the opto-insulated glass fibers in each discharge form a discharge bundle,

[0023]

[0020] - all together the discharge harnesses form the system bus,

[0024]

[0021] - via the system bus, already encoded in the binary positional system b-ary numbers enter and exit the processor.

[0025]

[0022] Further proposed is a computer including a processor, storage devices, input-output devices, an interface, in which the system bus is made in the form of a bundle consisting of bit bundles, each of which consists of individual opto-insulated glass fibers, along which optical pulses from laser beams move in a totality representing coded K-bit numbers in a binary positional number system, wherein the above-described method is used for the operation of the computer.

[0026]

[0023] Also proposed is an information and computing network that includes communication channels based on fiber-optic cables, data transmission equipment, hardware and software for switching nodes containing the above-mentioned set of devices for a computer and which functions according to the above-mentioned proposed method.

[0027]

[0024] DETAILED DESCRIPTION OF THE INVENTION

[0028]

[0025] The proposed method is carried out as follows.

[0029]

[0026] Each number in the system bus (address and command data) is represented as a conventional n-bit number in the b-ary positional number system, such that each digit in each n-th digit has one of the values ​​in the range from 0 to b - 1 and each digit is encoded by a K-bit number, but already in the binary positional number system and at the same time each digit in this number is represented by an optical metastable, monochrome laser beam with any wavelength in the range from 1 cm to 10 m.

[0030]

[0027] If the computer processor is implemented in electronic form on TTL or CMOS elements, i.e. not optical, then upon entering the processor, all optical signals in each individual opto-isolated fiber are preliminarily converted into logical "1" - if they are present, or logical "0" - if they are absent. And vice versa, upon exiting the processor, these signals in electronic form are converted into optical ones.

[0031]

[0028] Despite the fact that such a method of representing numbers and all buses complicates the design in quantitative terms of bundles, but at the same time simplifies the logistics of movements, transmission, reception of working optical signals due to parallelism and simultaneity both bit-by-bit and in buses (data, address and commands). And if we take into account that data can be represented by optical beams of any length and they do not need to be selected (separated by wavelength), then the speed of their transmission and transformations increases significantly. Now it is structurally determined from where and where and along which optical channel (fiberglass) the beam arrives in a given machine cycle.

[0032]

[0029] The advantages of this method can be considered using a specific example: when technically implementing the entire data bus in the amount of 16 bits, and each bit bundle of 8 separate opto-isolated glass fibers, then the maximum possible number in the data bus will be 256 to the power of 16 and equal to 3.4 * 10 to the power of 38. And since the maximum decimal number 256 in each bit is represented (encoded) by only one byte in the binary number system (2 to the power of 8 is equal to 256), then it will be written or read in an 8-bit memory register by one bit, and the entire 16-bit data bus by two bytes. It turns out that all numbers in this method are represented not by 64-bit binary numbers (one axis), but by 16 bits (along one axis) and one of the 256-bit levels - along the second axis.But since the number 256 is encoded, i.e. represented in binary code by one byte, the encoded word in the system bus is conventionally represented by two axes (2 bytes - along the digit and 1 byte - along the vertical Ь-ary axis) in the binary positional number system, i.e. suitable for processing in computers. Using the proposed method, the programmer will have access to each digit, i.e., the byte of information and, in general, all digits of the number. And with certain operations it (the number 3.4 * 10 to the power of 38) - will be addressed immediately by only a two-byte address. And in the case of processing information in the format of packed numbers or using the method of storing numbers in several cells with adjacent addresses, the entire word is addressed by one byte. This method of data representation is especially productive when processing character encoding tables in ASCII code or similar ones, as well as in network communication channels between individual computers and the control network server.

[0033]

[0030] The proposed inventions are intended to cover various modifications and changes within the scope of the essence and scope of the attached claims.

[0034]

[0031] SOURCES OF INFORMATION

[0035] 1. Article by Ugryumov E.P. Digital circuit engineering. Textbook for universities. St. Petersburg. "BHV- Petersburg. 2007. Paragraph 2.8. Adders. P. 114-1 16.

[0036] 2. Patent 6137 BY

[0037] 3. V.D. Sidorov, N.V. Strumpe. Computer hardware. Moscow, Publishing center "Academy". 2014. Pp. 41 - 45.

[0038] 4. Brusentsov N.P., Zhogolev E.A. Structure and algorithm of functioning of the small computing machine "SETUN" in the book "Computer engineering and issues of cybernetics". Publishing house Leningrad University 1971. issue 8, pp. 34-51

[0039] 5. S.A. Orlov. Organization of computers and systems. Petersburg. 2024, pp. 578-583

[0040] 6. V.D. Koldayev, S.A. Lupin. Computer Architecture. Moscow, ID "Forum" - Infra - M. 2014. Pp. 260 - 272.

Claims

FORMULA 1. A method for representing and processing data in a computer, characterized in that all information sent via the system bus to and from the processor is represented in the form of multi-bit numbers, with each digit represented in a positional number system, distinguished in that: - each number in the system bus is represented as a conditional n-digit number in the b-ary positional number system, - each digit in each digit, having one of the values ​​in the range from 0 to b-1, is represented by a K-digit number in the binary positional number system, - each digit in this number is represented by an optical laser beam, - each optical laser beam moves along a separate optically isolated glass fiber, - all together, the opto-isolated glass fibers in each discharge form a discharge bundle, - all together the discharge harnesses form a system bus, - via the system bus, already encoded in the binary positional system b - ary numbers enter and exit the processor.

2. A computer including a processor, storage devices, input-output devices, and an interface, characterized in that the system bus is made in the form of a bundle consisting of bit bundles, each of which consists of individual opto-insulated glass fibers, along which optical pulses from laser beams travel in the totality representing encoded K-bit numbers in a binary positional number system, wherein the method according to paragraph 1 is used for the operation of the computer.

3. An information and computing network that includes communication channels based on fiber-optic cables, data transmission equipment, hardware and software of switching nodes, characterized in that it contains a set of computer devices according to 2. and operates according to the method of paragraph.1, according to which all information sent via the system bus to the processor and from the processor is presented in the form of multi-bit numbers, wherein each bit is presented in the positional number system, wherein each number in the system bus is presented in the form of a conditional n-bit number in the b-ary positional number system, each digit in each bit, having one of the values ​​in the range from 0 to b-1, is represented by a K-bit number in the binary positional number system, each bit in this number is represented by an optical laser beam, each optical laser beam moves along a separate opto-isolated glass fiber, all together the opto-isolated glass fibers in each bit form a bit bundle, all together the bit bundles form a system bus, b-ary numbers already encoded in the binary positional system enter and exit the processor via the system bus.