Abstract: Static memory. Random access storage devices

The principle of memory homogeneity. Programs and data are stored in the same memory. Therefore, the computer does not distinguish between what is stored in a given memory cell - a number, text or command. You can perform the same actions on commands as on data. This opens up a whole range of possibilities. For example, a program can also be processed during its execution, which allows you to set rules for obtaining some of its parts in the program itself (this is how the execution of cycles and subroutines is organized in the program). Moreover, commands from one program can be obtained as results from the execution of another program. Translation methods are based on this principle - translating program text from a high-level programming language into the language of a specific machine.

The principle of program control. It follows from it that the program consists of a set of commands that are executed by the processor automatically one after another in a certain sequence.

Retrieving a program from memory is done using program counter. This processor register successively increases the address of the next command stored in it by the length of the command.

And since the program commands are located in memory one after another, a chain of commands is thereby organized from sequentially located memory cells.

If, after executing a command, you need to move not to the next one, but to some other one, use the commands conditional or unconditional transitions, which enter into the command counter the number of the memory cell containing the next command. Fetching instructions from memory stops after reaching and executing the instruction "stop".

Thus, the processor executes the program automatically, without human intervention.

3. The principle of targeting. Structurally, main memory consists of renumbered cells; Any cell is available to the processor at any time.

This implies the ability to name memory areas so that the values ​​stored in them can later be accessed or changed during program execution using the assigned names. Computers built on these principles are of the type von Neumann . But there are computers that are fundamentally different from von Neumann ones. For them, for example, maybe, i.e. they can operate without a “program counter” indicating the currently executing program command. To access any variable stored in memory, these computers you don't have to give her a name. Such computers are called non-von Neumann.

14. ARCHITECTURE AND STRUCTURE.

When considering computer devices, it is common to distinguish between their architecture and structure.

Architecture a computer is its description at some general level, including a description of user programming capabilities, command systems, addressing systems, memory organization, etc. The architecture determines the principles of operation, information connections and interconnection of the main logical nodes of a computer: processor, RAM, external storage and peripheral devices. The common architecture of different computers ensures their compatibility from the user's point of view.

Structure A computer is a set of its functional elements and connections between them. The elements can be a wide variety of devices - from the main logical nodes of a computer to the simplest circuits. The structure of a computer is graphically represented in the form of block diagrams, with the help of which you can describe the computer at any level of detail.

15. DISTINCTIVE FEATURES OF EACH OF THEM.

· Classical architecture (von Neumann architecture) - one arithmetic-logical unit (ALU), through which the data flow passes, and one control device (CU), through which the command flow - the program - passes. This single processor computer. This type of architecture also includes the architecture of a personal computer with common bus. All functional blocks here are interconnected by a common bus, also called system highway. Physically highway is a multi-wire line with sockets for connecting electronic circuits. The set of trunk wires is divided into separate groups: address bus, data bus and control bus.

Peripheral devices (printer, etc.) are connected to the computer hardware through special controllers - devices for controlling peripheral devices. Controller - a device that connects peripheral equipment or communication channels with the central processor, relieving the processor from directly controlling the operation of this equipment.

Multiprocessor architecture . Having multiple processors in a computer means that many data streams and many command streams can be organized in parallel. Thus, several fragments of one task can be executed in parallel.

Multi-machine computing system . Here several processors included in a computer system do not have common RAM, but they each have their own (local). Each computer in a multi-machine system has a classical architecture, and such a system is used quite widely. However, the effect of using such a computing system can only be obtained by solving problems that have a very special structure: it should be broken down into as many loosely coupled subtasks as there are computers in the system.

The speed advantage of multiprocessor and multi-machine computing systems over single-processor ones is obvious.

Parallel Processor Architecture . Here Several ALUs operate under the control of one control unit. This means that a lot of data can be processed by one program - that is, by one stream of commands. High performance of such an architecture can only be achieved on tasks in which the same computational operations are performed simultaneously on different data sets of the same type.

Modern cars often contain elements of various types of architectural solutions. There are also architectural solutions that are radically different from those discussed above.

16. CENTRAL PROCESSOR. TWO MAIN TYPES OF COMPUTER MEMORY.

The central processor generally contains:

• arithmetic-logical unit;
• data buses and address buses;
• registers;
• program counters;
• cache - very fast small memory (from 8 to 512 KB);
• mathematical floating point coprocessor.

Modern processors are implemented in the form microprocessors . Physically, a microprocessor is an integrated circuit - a thin rectangular wafer of crystalline silicon with an area of ​​only a few square millimeters, on which circuits are placed that implement all the functions of the processor.

The slab crystal is usually placed in a plastic or ceramic flat case and connected with gold wires to metal pins so that it can be attached to the computer's motherboard. A computer system may have several processors running in parallel; such systems are called

multiprocessor.

How memory works Computer memory is built from binary storage elements- bits , combined into groups of 8 bits, which are called(Memory units are the same as information units.) All bytes are numbered. The byte number is called it address.

Bytes can be combined into cells, also called words. Each computer has a specific word length - two, four or eight bytes. This does not preclude the use of memory cells of a different length (eg halfword, doubleword). Typically, one machine word can represent either one integer or one instruction. However, variable formats for presenting information are permitted. The breakdown of memory into words for four-byte computers is presented in the table:

Byte 0	Byte 1	Byte 2	Byte 3	Byte 4	Byte 5	Byte 6	Byte 7
HALF-WORD	HALF-WORD	HALF-WORD	HALF-WORD
WORD	WORD
DOUBLE WORD

Larger derived memory units are also widely used: Kilobyte, Megabyte, Gigabyte, and also, recently, Terabyte And Petabyte.

Modern computers have many different storage devices, which differ greatly in purpose, timing characteristics, volume of stored information and cost of storing the same amount of information. There are two main types of memory - internal And external.

17. MAIN COMPONENTS OF INTERNAL MEMORY. STATIC AND DYNAMIC MEMORY.

The internal memory includes RAM, cache memory And special memory.

1. RAM

RAM is only used for temporary storage of data and programs, because, when the machine turns off, everything that was in RAM is lost. Access to RAM elements straight- it means that Each byte of memory has its own individual address.

The amount of RAM is usually from 32 to 512 MB. For simple administrative tasks, 32 MB of RAM is sufficient, but complex computer design tasks may require 512 MB to 2 GB of RAM.

Typically RAM executed from SDRAM memory integrated circuits(synchronous dynamic RAM). Each information bit in SDRAM is stored as the electrical charge of a tiny capacitor formed in the structure of the semiconductor crystal. Due to leakage currents, such capacitors are quickly discharged, and they are periodically (about every 2 milliseconds) recharged by special devices. This process is called memory regeneration(Refresh Memory). SDRAM chips have capacity 16 - 256 Mbit and more. They are installed in housings and assembled into memory modules.

Most modern computers come with DIMM type modules(Dual-In-line Memory Module - a memory module with a two-row arrangement of chips). High-speed modules are used in computer systems on the most modern processors Rambus DRAM (RIMM) and DDR DRAM.

Memory modules are characterized by such parameters as volume-(16, 32, 64, 128, 256 or 512 MB), number of chips, nameplate frequency(100 or 133 MHz), data access time(6 or 7 nanoseconds) and number of contacts(72, 168 or 184). In 2001, production of memory modules began 1 GB and prototypes of modules for 2 GB. In 2009, 2 GB modules are common. Start of production of 4 GB modules.

2. Cache memory

The cache memory is controlled by a special device - controller, which, by analyzing the program being executed, tries to predict what data and commands the processor will most likely need in the near future, and pumps them into cache memory. In this case, it is possible to"hits" , so"misses"

. In case of a hit, that is, if the necessary data is pumped into the cache, it is retrieved from memory without delay. If the required information is not in the cache, the processor reads it directly from RAM. The ratio of hits to misses determines the effectiveness of caching. The cache memory is implemented on SRAM static memory chips (Static RAM), faster, more expensive and low-capacity than DRAM (SDRAM). Modern microprocessors have built-in cache memory , so-called first level cache 8, 16 or 32 KB in size. In addition, the computer's motherboard may have second level cache

with a capacity of 256, 512 KB and higher.

3. Special memory Special memory devices include permanent memory (ROM) reprogrammable read-only memory (Flash Memory) CMOS RAM memory , battery powered, video memory

and some other types of memory.

First of all, a program for controlling the operation of the processor itself is written into permanent memory. ROM contains programs for controlling the display, keyboard, printer, external memory, programs for starting and stopping the computer, and testing devices.

The most important permanent or Flash memory chip is the BIOS module. The role of the BIOS is twofold: on the one hand, it is an integral element of the hardware, and on the other hand, it is an important module of any operating system. (Basic Input/Output System - basic input/output system) - a set of programs designed for automatic testing of devices after turning on the computer and loading the operating system into RAM.

CMOS contents are changed by a special program Setup, located in the BIOS (English: Set-up - install, read “setup”).

Used to store graphic information , battery powered,.

There are many different types of RAM, but they can all be divided into two main subgroups - static memory (Static RAM) and dynamic memory (Dynamic RAM).

These two types of memory differ, first of all, in their fundamentally different technological implementation - SRAM will store recorded data until new ones are written or the power is turned off, and DRAM can store data only for a short time, after which the data must be restored (regenerated) , otherwise they will be lost.

Let's look at the advantages and disadvantages of SRAM and DRAM:

1. DRAM type memory, due to its technology, has a much higher data density than SRAM.

2. DRAM is much cheaper than SRAM,

3. but the latter is more productive and reliable, since it is always ready for reading.

STATIC RAM

In modern computers, SRAM is used as a second level cache and has a relatively small volume (usually 128...1024 KB). It is used in the cache precisely because very serious requirements are placed on it in terms of reliability and performance. The main memory of a computer is made up of dynamic memory chips.

Static memory is divided into synchronous and asynchronous. Asynchronous memory is no longer used in personal computers; it has been replaced by synchronous memory since the days of the 486 computers.

The use of static memory is not limited to cache memory in personal computers. Servers, routers, global networks, RAID arrays, switches - these are devices where high-speed SRAM is needed.

SRAM is a highly modifiable technology - there are many types, differing in electrical and architectural features. In conventional synchronous SRAM, there is a slight delay when the memory transitions from read mode to write mode.

Therefore, in 1997, several companies introduced their static RAM technologies without such a delay. These are ZBT (Zero-Bus Turnaround) SRAM technologies from IDT, and a similar NoBL (No Bus Latency) bus. DYNAMIC RAM (all memory except for the data segment - 64kb, stack memory - 16kb, own program body)

DRAM type memory is much more widespread in computing due to its two advantages over SRAM - low cost and data storage density. These two characteristics of dynamic memory compensate to some extent for its shortcomings - low performance and the need for constant data regeneration.

There are now about 25 varieties of DRAM, as memory manufacturers and developers try to keep up with advances in central processing units.

the main types of dynamic memory - from the old Conventional and FPM DRAM to the not yet implemented QDR, DDR SDRAM, RDRAM.

RAM has 3 sections:

• 640 kb. DOS - basic RAM
• 1MB Windows Core Modules – Top RAM
• the remaining modules are extended RAM

18. MEMORY MODULE DIMM. OTHER TYPES OF MEMORY MODULES.

Computer RAM is one of the most important elements of a computer, determining the performance and functionality of the entire system. RAM is represented by a certain number of RAM chips on the motherboard. If relatively recently RAM chips were connected through special sockets - connectors that made it possible to change individual chips without soldering, now the computer architecture provides for their placement on small module boards. Such memory modules are installed in special slots on the motherboard. One of the options for such a solution was SIMM modules (SIMM - single in-line memory modules).

Miniature SIMM modules, or simply SIMMs, are blocks of RAM of different capacities. SIMMs of 4, 8, 16, 32 and even 64 MB are widely used.

SIMMs come in two different types: 30 pin and 72 pin, where pin means the number of pins connected to a specialized RAM connector on the motherboard. At the same time, 30 pin and 72 pin SIMM are not interchangeable elements.

DIMM module appearance

DIMMs are most common as 168-pin modules that fit vertically into a socket and are secured with latches. SO DIMMs are widely used in portable devices - a type of small outline DIMM (SO - small outline), they are intended primarily for laptop computers.

Appearance of the RIMM module

Modules of the RIMM type are less common; such modules produce Direct RDRAM memory. They are represented by 168/184-pin rectangular boards, which must be installed only in pairs, and empty connectors on the motherboard are filled with special plugs. This is due to the design features of such modules.

19. EXTERNAL MEMORY. VARIETIES OF EXTERNAL MEMORY DEVICES.

External memory (ERAM) is designed for long-term storage of programs and data, and the integrity of its contents does not depend on whether the computer is turned on or off. Unlike RAM, external memory does not have a direct connection with the processor.

Information from the OSD to the processor and vice versa circulates approximately along the following chain:

VZU RAM ó Cache ó Processor

• The computer's external memory includes: drives for
• The computer's external memory includes: hard magnetic disks;
• The computer's external memory includes: flexible magnetic disks;
• The computer's external memory includes: CDs;
• The computer's external memory includes: Magneto-optical compact discs; magnetic tape

(streamers), etc.

1. Floppy disk drives
A floppy disk consists of a round polymer substrate coated on both sides with a magnetic oxide and placed in a plastic package with a cleaning coating applied to the inside surface. The packaging has radial slots on both sides through which the drive's read/write heads gain access to the disk. The method of recording binary information on a magnetic medium is called magnetic coding.

It lies in the fact that magnetic domains in the medium are aligned along paths in the direction of the applied magnetic field with their north and south poles. Usually set

There is a one-to-one correspondence between binary information and the orientation of magnetic domains. Information is recorded in concentric (paths tracks ), which are divided into sectors

. The number of tracks and sectors depends on the type and format of the floppy disk. A sector stores the minimum amount of information that can be written to or read from disk. The sector capacity is constant and amounts to 512 bytes. Currently the most widespread floppy disks with the following characteristics:

The floppy disk is installed in floppy disk drive(English) floppy-disk drive), is automatically recorded in it, after which the drive mechanism spins up to a rotation speed of 360 min -1. The floppy disk itself rotates in the drive, the magnetic heads remain motionless. The floppy disk rotates only when it is accessed. The drive is connected to the processor via floppy disk controller.

Recently, three-inch floppy disks have appeared that can store up to 3 GB information. They are manufactured using new technology Nano2 and require special hardware to read and write.

2. Hard disk drives

If floppy disks are a means of transferring data between computers, then hard drive - computer information warehouse.

Like a floppy disk, the working surfaces of platters are divided into circular concentric tracks, and the tracks into sectors. The read-write heads, along with their supporting structure and disks, are enclosed in a hermetically sealed housing called data module. When a data module is installed on a disk drive, it automatically connects to a system that pumps purified cooled air. Surface platter has magnetic coating only 1.1 microns thick and layer of lubricant to protect the head from damage when lowering and lifting while moving. When the platter rotates, a air layer, which provides an air cushion for hovering the head at a height of 0.5 microns above the surface of the disk.

Winchester drives have a very large capacity: from 10 to 100 GB. In modern models, the spindle speed (rotating shaft) is usually 7200 rpm, the average data search time is 9 ms, and the average data transfer speed is up to 60 MB/s. Unlike a floppy disk, a hard disk rotates continuously. All modern drives are equipped built-in cache(usually 2 MB), which significantly improves their performance. The hard drive is connected to the processor via hard drive controller.

4. CD drives

Here the storage medium is CD-ROM (Compact Disk Read-Only Memory - a compact disc from which you can only read).

The CD-ROM is a transparent polymer disk with a diameter of 12 cm and a thickness of 1.2 mm, on one side of which a reflective layer of aluminum is sprayed, protected from damage by a layer of transparent varnish. The coating thickness is several ten thousandths of a millimeter.

Information on the disk is presented as a sequence depressions(recesses in the disk) and projections(their level corresponds to the surface of the disk), located on a spiral track emerging from the area near the axis of the disk. For every inch (2.54 cm) of the radius of the disk there are 16 thousand turns of a spiral track. For comparison, only a few hundred tracks fit per inch radius on the surface of a hard drive. CD capacity reaches 780 MB. The information is written to the disc when it is manufactured and cannot be changed.

CD-ROMs have a high specific information capacity, which makes it possible to create on their basis help systems and educational complexes with a large illustrative base. One CD has the same information capacity as almost 500 floppy disks. Reading information from a CD-ROM occurs at a fairly high speed, although noticeably lower than the speed of hard disk drives. CD-ROMs are simple and easy to use, have a low unit cost of data storage, practically do not wear out, cannot be affected by viruses, and it is impossible to accidentally erase information from them.

Unlike magnetic disks, CDs do not have many ring tracks, but one - spiral, like gramophone records. In this regard, the angular speed of rotation of the disk is not constant. It decreases linearly as the laser reading head moves towards the edge of the disk.

To work with CD-ROM you need to connect it to your computer. CD-ROM drive(Fig. 2.9), which converts a sequence of indentations and protrusions on the surface of a CD-ROM into a sequence of binary signals. For this purpose it is used reading head with microlaser and LED. The depth of the depressions on the surface of the disk is equal to a quarter of the wavelength of laser light. If, in two successive cycles of reading information, the light beam of the laser head passes from the protrusion to the bottom of the depression or vice versa, the difference in the lengths of the light paths in these cycles changes to a half-wave, which causes an increase or decrease in the direct and reflected light from the disk hitting the LED together.

If the light path length does not change in successive read cycles, then the state of the LED does not change. As a result, the current through the LED produces a sequence of binary electrical signals corresponding to the combination of valleys and peaks on the trace.

The different lengths of the optical path of a light beam in two successive cycles of reading information correspond to binary units. Equal length corresponds to binary zeros.

Today, almost all personal computers have a CD-ROM drive. But many interactive multimedia programs are too large to fit on a single CD. CD-ROM technology is rapidly being replaced by DVD digital video disc technology.. These discs are the same size as regular CDs but can accommodate up to 17 GB of data, i.e. In terms of volume, they replace 20 standard CD-ROM drives. These discs are released on multimedia games and interactive videos excellent quality, allowing the viewer to view episodes from different camera angles, choose different ending options for the film, get acquainted with the biographies of the actors who starred, and enjoy excellent sound quality.

4. Magneto-optical CD drive DVD

4.7 17 50-hd dvd 200 blue ray

WARM drive(Write And Read Many times), allows you to write and read multiple times.

5. Magnetic tape drives (streamers)

Streamers allow you to record a huge amount of information onto a small magnetic tape cassette. The hardware compression tools built into the tape drive allow you to automatically compress information before recording it and restore it after reading it, which increases the amount of stored information.

The disadvantage of streamers is their relatively low speed of recording, searching and reading information.

1. Flash drive

Crystal on which information is recorded - 32GB

20. LIQUID CRYSTAL MONITORS. MONITORS BASED ON CRT

The computer video system consists of three components:

monitor(also called display);

video adapter;

software(video system drivers).

Video adapter sends beam brightness control signals and horizontal and vertical scanning signals to the monitor. Monitor converts these signals into visual images. A software process video images - perform signal encoding and decoding, coordinate transformations, image compression, etc.

The vast majority of monitors are designed based on cathode ray tube (CRT), and the principle of their operation is similar to the principle of operation of a TV. Monitors are alphanumeric and graphic, monochrome and color. Modern computers are usually equipped with color graphic monitors.

1. Monitor based on a cathode ray tube

The main display element is cathode-ray tube. Its front part, facing the viewer, is covered on the inside phosphor - a special substance capable of emitting light when hit by fast electrons.

The phosphor is applied in the form of sets of dots of three primary colors - red, green And blue . These colors are called primary because their combinations (in various proportions) can represent any color in the spectrum.

The sets of phosphor dots are arranged in triangular triads. The triad forms pixel- the point from which the image is formed (eng. pixel - picture element, picture element).

The distance between pixel centers is called monitor dot step. This distance significantly affects the clarity of the image. The smaller the step, the higher the clarity. Typically in color monitors the pitch is 0.24 mm. With this step, the human eye perceives the points of the triad as one point of a “complex” color.

On the opposite side of the tube there are three (according to the number of primary colors) electron guns. All three guns are “aimed” at the same pixel, but each of them emits a stream of electrons towards “its” phosphor point. In order for electrons to reach the screen unhindered, air is pumped out of the tube, and a high electrical voltage is created between the guns and the screen, accelerating the electrons. Placed in front of the screen in the path of the electrons mask- a thin metal plate with a large number of holes located opposite the phosphor points. The mask ensures that electron beams hit only the phosphor points of the corresponding color.

The magnitude of the electronic current of the guns and, consequently, the brightness of the pixels is controlled by the signal coming from the video adapter.

On the part of the flask where the electron guns are located, put deflection system monitor, which forces the electron beam to run through all the pixels one by one, line by line, from top to bottom, then return to the beginning of the top line, etc.

The number of lines displayed per second is called horizontal scanning frequency. And the frequency with which the image frames change is called frame rate. The latter should not be lower than 85 Hz, otherwise the image will be flicker.

2. LCD monitors

Increasingly used along with traditional CRT monitors. Liquid crystals- this is a special state of some organic substances in which they have fluidity and the ability to form spatial structures similar to crystalline ones. Liquid crystals can change their structure and light-optical properties under the influence of electrical voltage. By changing the orientation of groups of crystals using an electric field and using substances introduced into a liquid crystal solution that can emit light under the influence of an electric field, it is possible to create high-quality images that convey more than 15 million color shades.

Most LCD monitors use a thin film of liquid crystal sandwiched between two glass plates. Charges are transferred through the so-called passive matrix- a grid of invisible threads, horizontal and vertical, creating an image point at the intersection of the threads (somewhat blurred due to the fact that charges penetrate into adjacent areas of the liquid).

Active matrices Instead of threads, they use a transparent screen of transistors and provide a bright, virtually distortion-free image. The screen is divided into independent cells, each of which consists of four parts (for three primary colors and one reserve). The number of such cells according to the latitude and height of the screen is called screen resolution. Modern LCD monitors have a resolution of 642x480, 1280x1024 or 1024x768. Thus, the screen has from 1 to 5 million dots, each of which is controlled by its own transistor. In terms of compactness, such monitors have no equal. They take up 2 - 3 times less space than CRT monitors and are the same number of times lighter; consume much less electricity and do not emit electromagnetic waves that affect human health.

21. PRINTERS. PLOTTER. SCANNER

There are thousands of printer types. But there are three main types of printers: matrix, laser and inkjet.

· Dot matrix printers They use a combination of small pins that strike the ink ribbon, leaving an imprint of the symbol on the paper. Each character printed on the printer is formed from a series of 9, 18 or 24 needles formed in a vertical column. The disadvantages of these inexpensive printers are their noisy operation and poor print quality.

· Laser printers They work in much the same way as photocopiers. The computer forms an “image” of a page of text in its memory and transmits it to the printer. Information about the page is projected using a laser beam onto a rotating drum with a photosensitive coating that changes electrical properties depending on the light level.

After illumination, coloring powder is applied to the drum, which is under electrical voltage - toner, particles of which stick to illuminated areas of the drum surface. The printer uses a special hot roller to pull the paper under the drum; The toner is transferred to the paper and “fused” into it, leaving a durable, high-quality image. Colored Laser printers are still very expensive.

· Inkjet printers generate characters as a sequence ink dots. The printer's print head has tiny nozzles, through which quick-drying ink is sprayed onto the page. These printers are demanding on paper quality. Colored inkjet printers create colors by combining inks four primary colors - bright blue, purple, yellow and black.

The printer is connected to the computer via cable printer, one end of which is inserted with its connector into nest printer, and the other - in port computer printer. Port- this is a connector through which you can connect the computer processor to an external device.

Each printer must have its own driver- a program that is capable of translating (translating) standard computer printing commands into special commands required for each printer.

Plotters are used to produce complex design drawings, architectural plans, geographic and meteorological maps, and business diagrams. Plotters draw images using a pen.

Roller plotters scroll the paper under the pen, and flatbed plotters move the pen across the entire surface of the horizontally lying paper.

A plotter, just like a printer, definitely needs a special program - driver, allowing application programs to send instructions to it: raise and lower the pen, draw a line of a given thickness, etc.

If printers output information from a computer, then scanners, on the contrary, transfer information from paper documents to computer memory. Exist hand scanners, which are rolled over the surface of the document by hand, and flatbed scanners, in appearance reminiscent of copying machines.

Static random access memory(SRAM, static random access memory) -- Semiconductor random access memory in which each binary or ternary digit is stored in a positive feedback circuit that allows signal state to be maintained without the constant rewriting required in dynamic memory (DRAM). However, SRAM can only store data without overwriting it as long as there is power, that is, SRAM remains a volatile type of memory. Random access (RAM -- random access memory) -- the ability to select for writing/reading any of the bits (trites) (usually bytes (traits), depends on design features), in contrast to memory with sequential access (SAM -- sequental access memory).

Binary SRAM

Rice. 1.

A typical static binary memory cell (binary flip-flop) based on CMOS technology consists of two cross-connected (ring) inverters and key transistors to provide access to the cell (Fig. 1.). Often, to increase the packing density of elements on a chip, polysilicon resistors are used as a load. The disadvantage of this solution is the increase in static energy consumption.

The WL (Word Line) controls two access transistors. Lines BL and BL (Bit Line) - bit lines are used for both writing data and reading data.

Record. When a “0” is applied to the BL or BL line, parallel-connected transistor pairs (M5 and M1) and (M6 and M3) form 2OR logic circuits, the subsequent application of a “1” to the WL line opens the transistor M5 or M6, which leads to a corresponding switching of the trigger .

Reading. When “1” is applied to the WL line, transistors M5 and M6 open, the levels recorded in the trigger are set on the BL and BL lines and go to the reading circuits.

An eight-transistor binary SRAM cell is described in.

Switching flip-flops through access transistors is an implicit logical priority switching function, which explicitly, for binary flip-flops, is built on two-input logic elements 2OR-NOT or 2AND-NOT. The explicit switching cell circuit is a conventional RS flip-flop. With an explicit switching scheme, the read and write lines are separated, eliminating the need for access transistors (2 transistors per cell), but double-gate transistors are required in the cell itself.

Currently, an improved circuit has appeared (!) with feedback that can be switched off by the recording signal, which does not require load transistors and, accordingly, is free from high energy consumption during recording.

Trinity SRAM

Rice. 2. Ternary SRAM project on three-bit single-digit ternary flip-flops

One 2OR-NOT logic element consists of two two-gate transistors, three of six, plus three access transistors, for a total of nine transistors per one three-bit memory cell.

Advantages

· Fast access. SRAM is truly random access memory; accessing any memory cell at any time takes the same time.

· Simple circuit design - SRAM does not require complex controllers.

· Very low synchronization frequencies are possible, up to a complete stop of the clock pulses.

Flaws

· High Energy consumption.

· Low recording density (six elements per bit, instead of two for DRAM).

· As a result, the cost of a kilobyte of memory is high.

However, high power consumption is not a fundamental feature of SRAM; it is due to high exchange rates with this type of internal processor memory. Energy is consumed only when information in an SRAM cell changes.

Application

SRAM is used in microcontrollers and FPGAs, in which the amount of RAM is small (a few kilobytes), but low power consumption (due to the absence of a complex dynamic memory controller), predictable operating time of subroutines and debugging directly on the device are required.

In devices with a large amount of RAM, the working memory is executed as DRAM. SRAM is what makes registers and cache memory.

DRAM (dynamic random access memory)-- a type of volatile semiconductor random access memory (RAM), also the storage device most widely used as RAM in modern computers.

Physically, DRAM memory consists of cells created in semiconductor material, each of which can store a certain amount of data, from 1 to 4 bits. The set of cells of such memory forms a conditional “rectangle”, consisting of a certain number of rows and columns. One such “rectangle” is called a page, and the collection of pages is called a bank. The entire set of cells is conditionally divided into several areas.

As a storage device, DRAM memory is a module of various designs, consisting of an electrical board on which memory chips and a connector necessary to connect the module to the motherboard are located.

Rice. 3. Fig. 3.1

Physically, DRAM memory is a set of storage cells that consist of capacitors and transistors located inside semiconductor memory chips.

If there is no power supply to this type of memory, the capacitors are discharged and the memory is emptied (reset to zero). To maintain the required voltage on the plates of the capacitors of the cells and preserve their contents, they must be periodically recharged by applying voltage to them through switching transistor switches. This dynamic maintenance of capacitor charge is the fundamental operating principle of DRAM memory. The capacitors are charged when a one bit is written to the “cell”, and discharged when a zero bit needs to be written to the “cell”.

An important element of this type of memory is a sense amp connected to each of the columns of the “rectangle”. He, reacting to the weak flow of electrons rushing through the open transistors from the capacitor plates, reads the entire page. It is the page that is the minimum portion of exchange with dynamic memory, because data exchange with a single cell is impossible.

Regeneration

Unlike static memory type SRAM (English static random access memory), which is structurally more complex and more expensive type of memory and is used mainly in cache memory, DRAM memory is made on the basis of small capacitors, which quickly lose charge, therefore information must be updated at certain intervals to avoid data loss. This process is called memory regeneration. It is implemented by a special controller installed on the motherboard or on the central processor chip. During a time called the refresh step, an entire row of cells is rewritten into the DRAM, and after 8-64 ms, all memory rows are refreshed.

The process of memory regeneration in the classical version significantly slows down the operation of the system, since at this time data exchange with memory is impossible. Regeneration based on conventional row iteration is not used in modern DRAM types. There are several more economical options for this process - advanced, batch, distributed; The most economical is hidden (shadow) regeneration.

memory computer trigger cache

Triggers

Trigger (trigger system) is a class of electronic devices that have the ability to remain in one of two or more stable states for a long time and alternate them under the influence of external signals. Each trigger state is easily recognized by the output voltage value.

By the nature of their action, triggers belong to pulse devices - their active elements (transistors, lamps) operate in the switching mode, and the change of state lasts a very short time.

RAM collected on flip-flops is called static random access memory or simply static memory. The advantage of this type of memory is speed. Since the triggers are assembled on gates, and the gate delay time is very short, switching the trigger state occurs very quickly. This type of memory is not without its drawbacks. First, the group of transistors that make up a flip-flop is more expensive, even if they are etched in the millions on a single silicon substrate. In addition, a group of transistors takes up much more space because communication lines must be etched between the transistors that form the flip-flop. Used for ultra-fast RAM.

General information. Static memory (Static Random Access Memory - SRAM) is capable of storing data indefinitely without access (in the presence of supply voltage), i.e. V static mode. Static memory cells are built on elements with two stable states (bistable cells or flip-flops). Compared to dynamic capacitive memory elements, they are easier to manage and do not require regeneration, but are more complex in circuitry and take up more space on the chip. The performance and power consumption of static memory are determined by the manufacturing technology and circuit design of the storage cells. The most cost-effective CMOS Memory is suitable for long-term data storage when powered by a low-power battery. It is used in the configuration memory of personal computers. CMOS memory access time is more than 100 ns. The fastest static memory has access times of several nanoseconds (and even tenths of a nanosecond). Such memory is capable of operating at the system bus frequency together with the processor, without requiring wait cycles from it.

The typical memory capacity of modern SRAM chips reaches 1 Mbit or more. Exist three varieties static memory chips: Async SRAM, Sync Burst SRAM and Pipelined Burst SRAM. The relatively high specific cost of data storage at a low packaging density does not allow the use of SRAM as the main memory of computers.

To avoid cost increases, computers are equipped with a small amount of high-speed SRAM, which is used as cache. Cache memory is capable of operating at clock speeds close to or equal to processor clock speeds. Therefore, it is directly used by the processor when reading and writing, which reduces the number of downtimes and increases the performance of the computer as a whole. The cache controller anticipates the processor's data needs and preloads the required data into high-speed cache memory. When the processor issues a memory address, the data is transferred not from slow RAM, but from the cache.

To reduce latency and processor downtime when reading data from low-speed RAM, modern computers provide up to three cache levels. In this case, the cache memory of the first and second levels can be located on the same chip with the processor. The use of synchronous operation with the processor and pipelined batch mode increases the speed and efficiency of the cache memory. The capabilities and effectiveness of the cache memory are determined by the controller, which is located in the system logic chips (usually North Bridge) or on the processor board.

Thus, the main features of static RAM include:

• ability when the computer is on store indefinitely data (information) in the absence of requests. This capability is provided by bistable bistable memory cells, which are implemented on bipolar or CMOS structures;
• relatively high performance microcircuits based on bipolar structures (access time is a few nanoseconds), allowing operation synchronously with processors at frequencies above 500 MHz;
• low power consumption CMOS chips that provide long-term storage of input/output system (BIOS) parameters;
• relatively large dimensions microcircuits and high price, which is associated with a large number of transistors and their clustered placement (clusters of six transistors are used);
• typical Memory SRAM chips reach 1 Mbit or more;
• main application area– cache memory and computer configuration memory.

Static memory - SRAM (Static Random Access Memory), as its name suggests, is capable of storing information in a static mode - that is, for an indefinitely long time in the absence of access (but in the presence of supply voltage). Static memory cells are implemented on flip-flops - elements with two stable states. Compared to dynamic memory, these cells are more complex and take up more chip space, but they are easier to manage and do not require regeneration. The performance and power consumption of static memory are determined by the manufacturing technology and circuit design of the storage cells.
The most economical static memory CMOS (or CMOS Memory) is at the same time the slowest memory of this type, has an access time of more than 100 nanoseconds, but is suitable for long-term storage of information when powered by a low-power battery. CMOS memory is used in personal computers to store configuration data and implement an internal clock.
The fastest static memory has an access time of a few nanoseconds, which allows it to operate at the processor's system bus speed without requiring processor wait cycles. The relatively high specific cost of storing information and high power consumption with a low packing density of elements does not allow the use of SRAM as computer RAM.
Static memory devices (SRAM) have the advantage over dynamic ones that their access time is almost equal to the write or read cycle time. Made using the same technology as the processor, static memory has high performance. The main limitation in using static memory is cost. With equal capacity to dynamic memory, static memory is approximately four times more expensive. Therefore, this type of memory has become widespread in high-performance systems as an external (relative to the processor) cache memory. The price/performance ratio in these systems does not play such a significant role. However, with the advent of high-capacity static memory chips and its reduction in cost, the existing stereotype of using memory circuits will change and computer manufacturers may replace dynamic memory with static memory, while in the meantime static memory elements are used in RAM as a fast pipeline buffer for data preparation to output data to the bus every cycle of the system bus.
Structure of a static memory chip
The memory element in static RAM is a trigger made of transistors. The structure of a static memory chip (Fig. 1.) includes a storage matrix containing M x N memory elements.

Dynamic memory (DRAM) is a type of random access memory used in computing devices, most notably PCs. DRAM stores each bit of data in a separate passive electronic component that resides inside an integrated circuit board. Every electrical component has two value states in one bit, called 0 and 1. It must be updated frequently, otherwise the information disappears. DRAM has one capacitor and one transistor per bit, unlike static random access memory (SRAM) which requires 6 transistors. The capacitors and transistors used are extremely small. There are millions of capacitors and transistors that fit into a single memory chip.

As a form of memory technology, dynamic RAM memory arose from the developments of early microprocessors and related integrated circuit developments. In the mid-1960s, they began to appear in some modern electronic products that previously used a form of magnetic memory in the form of one small ferrite toroid for each element. Naturally, this "main" memory was very expensive, and integrated versions were more attractive in the long run.

The idea of ​​DRAM technology appeared relatively early in the semiconductor integrated circuit timeline. An early form was used in Toshiba's calculator, which was released in 1966 from a discrete component, and then the idea was patented two years later. The next stage of technology development occurred in 1969, when Honeywell, which had entered the computer market, asked Intel to make dynamic memory using three transistor cell ideas. The resulting DRAM IC was called the Intel 1102 and appeared in early 1970. However, the device had several problems, after which Intel developed new technology that worked more reliably.

The resulting new device appeared in late 1970 and was called the Intel 1103. The technology advanced even further when MOSTEK released its MK4096 in 1973. As the part number indicates, the device had a capacity of 4 k. Its main advantage was that it included a multiplexed row and column approach. This new approach made it possible to fit into packages with fewer contacts. As a result, the cost advantage increases over previous approaches with each increase in memory capacity.

This allowed MOSTEK technology to gain more than 75% of the global market share. MOSTEK eventually lost out to the Japanese manufacturers because they were able to produce better devices at a lower price.

DRAM is dynamic memory and SRAM is static memory. The DRAM chips on the board are updated every few milliseconds. This is done by rewriting the data to the module. The chips that need updating are volatile memory. DRAM directly accesses memory, retains memory for a short period, and loses its data when power is removed.

SRAM is one that is static and does not need to be updated. Since it runs much faster, it is used in registers and cache memory. SRAM stores data and operates at higher speeds than motherboard DRAM because it is much cheaper to manufacture.

DRAM is a type of semiconductor memory that a system designer can use when building a computer. Alternative memory options include static RAM (SRAM), electrically erasable programmable read-only memory (EEPROM), NOR Flash, and NAND Flash. Many systems use more than one type of memory.

Types of printed circuit boards and reading systems

The three main types of printed circuit boards that contain memory chips are dual embedded memory modules (DIMMs), single-line memory modules (SIMMs), and Rambus in-line memory modules (RIMMs).

Today, most motherboards use DIMMs. The module refresh rate for DRAM is every few milliseconds (1/1000 of a second). This update is performed by the memory controller located on the motherboard chipset. Since refresh logic is used to automatically refresh, the DRAM board is quite complex.

There are various systems used for updating, but all methods require a counter to keep track of the row that needs to be updated as follows. DRAM cells are organized as a square array of capacitors, typically 1024 by 1024 cells. When a cell is in the read state, the entire row is read and the update is written back. When in the "write" state, the entire row is "read", one value is changed, and then the entire row is rewritten.

Depending on the system, there are DRAM chips that contain the counter while other systems rely on peripheral update logic. Access time is around 60 nanoseconds, while SRAM can reach 10 nanoseconds. In addition, the cycle time of DRAM is much longer than that of SRAM. The cycle time is shorter because it does not have to stop between calls and updates.

DRAM is the successor to SRAM. Memory designers reduced the number of elements per bit and eliminated differential bit lines to save chip area for DRAM creation. As a result, it is cheaper to produce than SRAM. But SRAM retains some advantages over DRAM. Comparison of static and dynamic memory:

1. SRAM does not need to be refreshed because it works by switching current flow in one of two directions instead of holding charge in a storage location.
2. It is typically used for cache memory, which can be accessed faster than DRAM.
3. SRAM is capable of reading and writing byte bits and is faster at reading and writing than DRAM, which writes data at the byte level and reads at the multi-byte page level.
4. Differences in power are determined depending on whether the system is in active or sleep mode. DRAM requires less power than SRAM when active, but SRAM consumes significantly less power when idle.

There are many types or interfaces for communicating with DRAM. These include fast page mode (FPM DRAM), extended data from DRAM (EDO RAM), and synchronous DRAM (SDRAM). SDRAM is a generic name for types of DRAM that are clocked by a microprocessor. These include single data rate SDRAM (SDR), double data rate SDRAM (DDR), DDR2 SDRAM, DDR3 SDRAM, and DDR4 SDRAM.

How RAM works

Dynamic memory devices have MOS technology underlying their design, manufacturing and operation. Looking at how DRAM memory works, you can see that basic RAM or DRAM uses a capacitor to store each bit of data and a transfer device - a MOSFET - that acts as a switch.

The level of charge on the capacitor of a memory cell determines whether that particular bit is a logic "1" or "0" - the presence of charge in the capacitor indicates a logic "1", and the absence of charge indicates a logic "0". Dynamic memory allocation RAM has a specific format that allows it to be tightly packed onto a silicon chip, making it very cheap. Two lines are connected to each dynamic RAM cell - a Word line (W/L) and a Bit line (B/L), so that the desired cell within the matrix can read or write data.

Base cell

The basic memory cell shown would be one of many thousands or millions of such cells in a complete memory chip. They can have a capacity of 256 Mbit or more. To improve write and read capabilities and speed, memory is dynamically allocated and divided into submatrices. Having multiple subarrays shortens words and bit strings, and this reduces access time to individual cells. For example, 256 Mbit DRAM can be divided into 16 smaller 16 Mbit arrays.

The linear ones control the input of the transmission lines, while the bit bins are connected to the FET channel and ultimately connected to the gain amplifiers. There are two ways to organize bit strings:

1. Stacked bit lines. You can think of a pair of adjacent bit lines as one bit line folded in half, with the connection on the slot connected to a shared amplifier. This format provides additional noise immunity, but at the expense of compactness.
2. Open bit lines. In this configuration, the lines are placed between two subarrays, thereby connecting each signal amplifier to one bit line in each array. This offers a more compact solution than folded bitlines at the expense of noise immunity.

One of the problems with this circuit is that the capacitors do not hold their charge indefinitely because there is some leakage across the capacitor. It would be unacceptable for the memory to lose its data, and to overcome this problem, the data is updated periodically. The data is read and written, and this ensures that any leakage is overcome and the data is recovered.

One of the key elements of DRAM memory is the fact that the data is updated periodically. Typically, manufacturers specify that each row should be updated every 64 ms. This time interval complies with JEDEC standards for dynamic RAM refresh periods.

There are many ways in which you can update. Some processor systems update each row together every 64 ms. Other systems update one row at a time, but this has the disadvantage that with large memories the update rate becomes very fast. Other systems, especially real-time systems where speed is important, take the approach of having part of the semiconductor memory simultaneously dependent on an external timer that controls the operation of the rest of the system. Thus, it does not interfere with the operation of the system.

Regardless of which method is used, it is necessary for the counter to be able to keep track of the next line in the DRAM that needs to be updated. Some chips include a counter, otherwise an additional device must be added for this purpose. It may seem that the refresh circuits required for DRAM memory would add complexity to the overall memory design and make it more expensive. However, it has been found that the additional circuitry is not a serious problem if it can be integrated into the memory chip. And it's also found to be much cheaper and have much higher capacity than its other main contender, static RAM (SRAM).

Signal to noise ratio

As memories increase in size, the issue of signal-to-noise ratio becomes very important because it can cause data corruption problems. This depends on the ratio of the capacitance of the storage capacitor in the DRAM memory to the capacitance of the Word or bit line that is dumped when the cell is accessed. As the bit density per chip increases, the ratio worsens as the cell area decreases due to more cells being added to the bit line.

For this reason, it is important to store both high voltage on the capacitor and increase the DRAM storage capacity for given areas as much as possible. This is very important because the sensitivity of the small charge on the capacitor of a memory cell is one of the most challenging areas of DRAM memory chip design. As a result of this, some complex circuits were included in memory chips.

DRAM memory chips are widely used and the technology is very well proven. And memory chips and plugins are available to expand the memory of computers and many other devices. Although DRAM has its disadvantages, it is still widely used as it offers many advantages in terms of cost and satisfactory speed, it is not the fastest, but is still much faster than some other types of memory.

There are several types within the DRAM memory family, including asynchronous, synchronous, EDO, BEDO, FPM, and others. Besides the type of memory technology, it can also be contained in several types of IC packages. DRAM is also available in module formats and there are several types of memory modules including DIMMs, SIMMs, RIMMs, etc. Thus, it is necessary to have an understanding of all the different types of DRAM and formats in which memory can be obtained, installed and used.

When studying memory technology itself, there is a wide variety of different types of DRAM. Asynchronous DRAM is the basic type on which all other types are based. Asynchronous ones have connections for power, address inputs, and bidirectional data lines. Although this type of DRAM is asynchronous, the system is run by the memory controller, which is synchronized, and this limits the speed of the system to multiply the clock speed. However, DRAM operation itself is not synchronous.

Memory allocation

Dynamic memory allocation is the process by which computer programs and services are assigned physical or virtual memory space. In fact, it is the process of reserving a partial or complete portion of computer memory for the execution of programs and processes. Memory allocation is achieved through a process known as memory management through the operating system and software applications.

Dynamic memory allocation has two main types:

1. Static memory allocation, memory is allocated to a program at compile time.
2. Dynamic Memory Allocation,Programs are allocated memory at runtime.

The process of memory allocation is very similar to the management of physical and virtual memory. Programs and services are assigned specific memory according to their requirements during execution. Once a program has completed its work or is idle, the memory is freed and assigned to another program or merged into primary memory.

Optimizing Memory Usage

The arduino's dynamic memory is in the form of flash. Where the program itself is stored and cannot be changed except when the user loads a new program, called a "sketch", from the computer, and retains what is downloaded even if the power is turned off. When checking or loading a sketch, the PC will report in the window how much flash there is and how much has been used if “detailed mode” is enabled in the settings.

Every time a new thumbnail is loaded, it overwrites the old one. The Arduino only has one program at a time, and when power is supplied to the Arduino, the program runs forever. Most modern Arduinos have about 32K flash memory, which is quite small and limits the size of the programs (sketchbooks) you can download. But SRAM is the real limit for many things. The user really needs to be careful in planning to minimize what really needs to be kept. And if they try to use too much, the Arduino simply won’t work. The user will not even be able to perform the most minimal debugging actions until the PC is rebooted.

SRAM is the most valuable memory commodity on Arduino. Although SRAM flaws are probably the most common memory problems on Arduino. They are difficult to diagnose. If the program fails inexplicably, there is a good chance that the user crashed the stack due to not having enough SRAM. There are a number of things that can be done to reduce SRAM usage:

1. Remove unused variables.
2. Reserve lines.
3. Move permanent data to PROGMEM.
4. Reducing buffer sizes.
5. Reducing oversized variables.

Any variable that the user defines either at the top of the program, inside a function, or even on the fly in something like a for loop will likely use SRAM, although some variables are never stored in SRAM. Every time the Arduino is started via power on or reset, all of its variables are reinitialized to default and it needs to relearn the environment it is working with.

Working with dynamic memory - an important important aspect to consider when designing a system. In fact, there is a third type of memory - EEPROM, which can be written to and will be retained in the event of a power failure. Arduino can write 300 EEPROM per second, if the user is not careful, then theoretically this speed can destroy a memory cell in 5 minutes, and the entire EEPROM in two days.