What is disk and flash memory. Flash memory. Operating principle

A new word has appeared in users' everyday life - “flash drive”. Most people know for sure that this device is used in digital cameras, and is also intended for transferring videos and music. But this is not a complete list of functions that a flash drive performs. This device is indispensable in the work of any owner of not only a computer, but also all kinds of electronics of the 21st century. The topic of this article is flash memory, its characteristics, types, prices.

Plunging into history

All the giants of the IT industry are engaged in rewriting history, establishing their authorship over various inventions. The famous American company Intel did the same and took credit for the invention of flash memory. However, the technology and production of the world's first device belong to the Japanese giant Toshiba, which presented its discovery to the world back in 1984. The name “flash memory” was also given to the device by the Japanese, and not by chance. The process of erasing information on flash memory is vaguely reminiscent of a photo flash.

Less than a few years have passed since the invention, and the global giants of the IT industry quickly found application for the new invention, putting production on the assembly line.

Not all memory is flash

Taking a shallow dive into the world of physics, you can find out that memory comes in several types.

1. RAM, which works on the principle of “electrical capacitance”. Millions of capacitors, holding a charge in RAM, are storehouses of information. When the power supply to the device is turned off, the capacitors are discharged, losing information forever.
2. Permanent memory. Information on the medium is stored through physical or chemical influence. An example is an optical DVD, on which information is written by burning microscopic holes on the surface of the plastic with a laser.
3. Conditionally permanent non-volatile memory. This includes flash memory, magnetic hard drives, floppy disks, video tape and other media that can hold a magnetic or electrical charge in the absence of a constant source of electricity.

Applications of Flash Memory

For 20th century technology, devices such as memory cards and USB flash drives were sufficient for the invention. But in the 21st century there was a boom in storage media with flash technology. First of all, all mobile phones, tablets, multimedia players and digital devices acquired flash memory. Later, not a single interactive toy for children could exist without flash memory. Technology didn't stop there. Every day new devices appear, equipped with this wonderful type of memory. Take at least a flashlight for a policeman. Thanks to the presence of flash memory, a human rights activist can select the desired focus and brightness of the beam from the saved settings.

How many device manufacturers

On the market you can see that the required flash memory is presented by several manufacturers. Having almost identical characteristics, drives differ significantly in price. Is the most expensive purchase really the best? Not always! Often the buyer has to overpay for the brand, service and warranty.

There are only a few factories in the world that produce flash memory modules. These modules are bought up by IT industry giants, who create a beautiful case and sell the drive on their own behalf. The only difference is the speed of the devices, depending on the capabilities of the flash memory. Whether the memory is fast or not is decided by the manufacturer.

About prices for flash devices

To anyone who has independently decided to purchase flash memory, prices on the market may seem strange. Drives of the same capacity from two unfamiliar brands have a large difference in cost. What's the matter? There is a set of requirements for flash cards, thanks to which the manufacturer is obliged to determine the class of the device and make markings on the product body. Often in the store you can find devices that have no markings, only the logo of the manufacturer. The prices for such memory cards are very low, and the seller claims high performance of the device. Reviews from experts on the pages of respected computer publications recommend refraining from purchasing unmarked devices, as they are counterfeit or imported into the country illegally.

What you need to know about flash drive labels

Since we are talking about drive labeling, when purchasing, you need to pay attention to the numbers and inscriptions indicated on the housing of flash devices.

1. The name of the manufacturing company or its logo must be present.
2. The amount of flash memory must be written on the media.
3. The class of the flash device must be indicated on the case. Manufacturers of USB modules often indicate the class on the product packaging, which is not prohibited by law.

On sale you can find flash memory cards without markings, but with a long number that small print stamped on the device body. Thus, the manufacturer indicates the batch number by which the buyer can find the device on the Internet and get acquainted with its technical characteristics.

Speed ​​is proportional to price, but not to efficiency

The higher the class of flash memory, the higher its write speed, and the higher the price. Is it worth buying the fastest memory?

1. Zero class. Write speed of at least 0.6 MB per second. You can buy it in stores without seeing the lack of labeling. Suitable for storing documentation.
2. Classes 2 and 4, with write speeds of 2 and 4 MB per second, respectively, also belong to the office section and are intended for storing and transferring documentation.
3. Sixth and eighth grades with speeds of 6 and 8 MB per second will be of interest to all customers who work with photos, music, and video. These types of flash memory unlock the potential for multimedia applications.
4. Class ten and above, including Ultra, show write speeds of over 10 MB per second. Used in working with multimedia, as additional drives for workstations, used as RAM. Where the speed of reading and writing to the storage medium is critical.

Serious brands such as Pretec and Corsair make high-speed devices with write capabilities of about 25 MB per second, labeling them eighth or tenth class. The price of the modules is very high, but in the IT world such brands are highly respected by users.

What are the different amounts of flash memory

Another criterion on which the price of a drive depends is the amount of flash memory. Even though technology does not stand still, there are still some limits. When it is necessary to change the technical process to increase the amount of memory, a dilemma arises - while maintaining a low price, stop at the achieved result or develop further, looking for rich buyers. There has been some calm in the world - buyers are offered to buy memory cards with a maximum capacity of 64 gigabytes; if you really want, you can order 128 GB and 256 GB, but for this you will have to fork out a lot. It is unknown how long it will take for the transition to new technologies and the availability of large-capacity cards on the market, but one thing is known - 64 GB is quite enough to satisfy any task of the average user.

A miracle beast with a great future

There is another interesting device that uses flash memory in its operation - SSD drive. Along with the recording volume and speed, the authority of the manufacturer is critical for the device, which provides the product with a control controller and specialized firmware that controls the entire device. One mistake by the manufacturer and the device could end up in the trash. Everything is complicated, expensive and very serious, but the SSD drive is the future. Direct competitor hard drives computers that work using magnetism. It is resistant to shaking, temperature and operates silently. The day is not far off when magnetic hard drives will share space in the closet with ball mice, giving way to 21st century technology.

How to save money on upgrading your computer

Owners of old computers and laptops have often heard from service specialists about the reasons for the low speed of the device. There is not enough RAM, which has long been discontinued. The specialist, looking into the eyes of the computer owner, convinces that the only way out of the situation is to purchase modern computer. After 5 years, the same specialist will come and once again prove that there are no solutions other than buying a new computer. This is how the world is built. A world for people who are not interested in knowledge of IT technologies.

RAM flash memory will solve the problem once and for all at minimal cost to the user. Just download a program called Ready Boost from the Internet and study system requirements to the drive. And only then purchase the necessary flash memory device in the store. Connect the drive to your computer or laptop, run the program and enjoy life. It’s so nice to independently increase the performance of your computer without capital investments.

Which brand should you prefer?

Due to the large number of manufacturers, it is very difficult to decide who to give preference to. Experts recommend making a list of requirements for the drive, and then choosing a brand.

1. The purpose of use allows you to identify the required class of device.
2. Convenience and appearance will tell you what a flash drive should look like. For example, for a car radio, you should pay attention to a small drive so as not to accidentally break it during use.

Having found several required options, ask the seller how issues are resolved when the device breaks down, whether there is warranty replacement. Flash memory is a consumable item and cannot be repaired - you need to know this before purchasing. Positive Feedback deserve the manufacturers Corsair, Kingston, OCZ, Pretec, Silicon Power, Transcend and IBM.

How to protect yourself from losing information from a flash drive

Like any storage medium, a memory card is subject to influence external factors, which all users of flash devices need to know about and worry about the safety of their information.

1. Physical failure of modules. Plastic flash cards are very easy to break and impossible to restore, so when purchasing you need to pay attention to metal flash cards or use them very carefully.
2. Moisture can destroy the drive. It is worth paying attention to waterproof media if there is a possibility of water getting into the memory.
3. Flash memory infection by viruses. Sometimes it turns out to be quite difficult to recover information, so you should pay attention to devices that have physical write protection in the form of a switch - this is guaranteed not to give viruses a single chance.

Having found out the principle of operation, types, characteristics, prices and design of flash memory, you need to entrust your choice to professionals.

1. Experts recommend giving preference to proven brands. To do this, just turn to popular sources of information and read reviews about the product. Any self-respecting manufacturer on the Internet has its own website. This is where it is worth visiting to get an idea of ​​the company.
2. You should not trust your choice to Chinese counterfeits, which are offered on the market at a very low price. If there are no other options, be sure to ask the seller to demonstrate the operation of the media before purchasing. Normal device formatting Windows environment allows you to determine the health of flash memory.
3. Preference should be given to fast devices that have a tenth class rating. Because there are often situations when time is at a premium. Then flash memory will become universal for the user for any device.
4. When buying memory cards for digital equipment, you should worry about the capabilities of reading data on a computer. For this, there are all kinds of adapters, which are often offered for purchase along with flash memory.

Theoretical part

What kind of memory is there?

The only thing that, perhaps, can unite all these types of memory is more or less the same operating principle. There is some two-dimensional or three-dimensional matrix that is filled with 0s and 1s in approximately this way and from which we can subsequently either read these values ​​or replace them, i.e. all this is a direct analogue of its predecessor - memory on ferrite rings.

What is flash memory and what types does it come in (NOR and NAND)?

Schematic representation of a floating gate transistor.

So how does this engineering marvel work? This is described together with some physical formulas. In short, between the control gate and the channel through which current flows from source to drain, we place the same floating gate, surrounded by a thin layer of dielectric. As a result, when current flows through such a “modified” field-effect transistor, some high-energy electrons tunnel through the dielectric and end up inside the floating gate. It is clear that while the electrons were tunneling and wandering inside this gate, they lost some of their energy and practically cannot return back.

NB:"practically" - keyword, because without rewriting, without updating cells at least once every few years, Flash is “reset to zero” in the same way as RAM, after turning off the computer.

Again we have a two-dimensional array that needs to be filled with 0s and 1s. Since it takes quite a bit to accumulate charge on the floating gate for a long time, then in the case of RAM a different solution is applied. The memory cell consists of a capacitor and a conventional field-effect transistor. Moreover, the capacitor itself has, on the one hand, a primitive physical device, but, on the other hand, it is non-trivially implemented in hardware:

RAM cell design.

Again, ixbt has a good one dedicated to DRAM and SDRAM memory. It is, of course, not so fresh, but the fundamental points are described very well.

The only question that torments me is: can DRAM have a multi-level cell, like flash? It seems like yes, but still...

Practical part

Flash

The main elements of a USB Flash drive: 1. USB connector, 2. controller, 3. PCB-multilayer printed circuit board, 4. NAND memory module, 5. quartz reference frequency oscillator, 6. LED indicator (now, however, on many flash drives do not have it), 7. write protection switch (similarly, it is missing on many flash drives), 8. space for an additional memory chip.

Let's go from simple to complex. Crystal oscillator(more about the principle of operation). To my deep regret, during the polishing the quartz plate itself disappeared, so we can only admire the body.

Crystal oscillator housing

By chance, in the meantime, I found what the reinforcing fiber inside the PCB looks like and the balls that make up the PCB for the most part. By the way, the fibers are still laid with twisting, this is clearly visible in the top image:

Reinforcing fiber inside the PCB (red arrows indicate fibers perpendicular to the cut), which makes up the bulk of the PCB

And here is the first important part of the flash drive - the controller:

Controller. The top image was obtained by combining several SEM micrographs

To be honest, I didn’t quite understand the idea of ​​the engineers who placed some additional conductors in the chip itself. Maybe it's from the point of view technological process easier and cheaper to do.

After processing this picture, I shouted: “Yayyyyyyyy!” and ran around the room. So, we present to your attention the 500 nm technological process in all its glory with perfectly drawn boundaries of the drain, source, control gate, and even the contacts are preserved in relative integrity:

"Ide!" microelectronics – 500 nm controller technology with beautifully drawn individual drains (Drain), sources (Source) and control gates (Gate)

Now let's move on to dessert - memory chips. Let's start with the contacts that literally feed this memory. In addition to the main one (the “thickest” contact in the picture), there are also many small ones. By the way, "fat"< 2 диаметров человеческого волоса, так что всё в мире относительно:

SEM images of the contacts powering the memory chip

If we talk about memory itself, then success awaits us here too. We were able to photograph individual blocks, the boundaries of which are indicated by arrows. Looking at the image with maximum magnification, try to strain your gaze, this contrast is really difficult to discern, but it is there in the image (for clarity, I marked a separate cell with lines):

Memory cells 1. Block boundaries are marked with arrows. Lines indicate individual cells

At first it seemed to me like an image artifact, but after processing all the photos of the house, I realized that these are either control gates elongated along the vertical axis in an SLC cell, or these are several cells assembled in an MLC. Although I mentioned MLC above, this is still a question. For reference, the "thickness" of a cell (i.e. the distance between two light dots in the bottom image) about 60 nm.

In order not to dissemble, here are similar photos from the other half of the flash drive. A completely similar picture:

Memory cells 2. Block boundaries are highlighted with arrows. Lines indicate individual cells

Of course, the chip itself is not just a set of such memory cells; there are some other structures inside it, the identity of which I could not determine:

Other structures inside NAND memory chips
 

DRAM

However, according to tradition, let's start with contacts. It was nice to see what a BGA “chip” looks like and what the soldering itself is like:

"Chipped" BGA solders

And now it’s time to shout “Ide!” for the second time, since we managed to see individual solid-state capacitors - concentric circles in the image, marked with arrows. They are the ones who store our data while the computer is running in the form of a charge on their plates. Judging by the photographs, the dimensions of such a capacitor are about 300 nm in width and about 100 nm in thickness.

Due to the fact that the chip is cut at an angle, some capacitors are cut neatly in the middle, while others have only the “sides” cut off:

DRAM memory at its finest

If anyone doubts that these structures are capacitors, then you can look at a more “professional” photo (though without a scale mark).

The only point that confused me is that the capacitors are located in 2 rows (lower left photo), i.e. It turns out that there are 2 bits of information per cell. As mentioned above, information on multibit recording is available, but to what extent this technology is applicable and used in modern industry remains questionable to me.

Of course, in addition to the memory cells themselves, there are also some auxiliary structures inside the module, the purpose of which I can only guess:

Other structures inside a DRAM memory chip

Afterword

Unfortunately, it was not possible to find a large number of videos on the topic of Flash and RAM production, so you will have to be content with only assembling USB Flash drives:

P.S.: Once again, Happy New Year of the Black Water Dragon everyone!!!
It turns out strange: I wanted to write an article about Flash one of the first, but fate decreed otherwise. Fingers crossed, let's hope that the next at least 2 articles (about biological objects and displays) will be published in early 2012. In the meantime, the seed is carbon tape:

Carbon tape on which the studied samples were fixed. I think regular tape looks similar.

Flash memory

Story

Flash memory was discovered by Fujio Masuoka while he was working at Toshiba in 1984. The name "flash" was also coined at Toshiba by Fuji's colleague, Shoji Ariizumi, because the process of erasing the contents of memory reminded him of a flash. Masuoka presented his design at the IEEE 1984 International Electron Devices Meeting (IEDM), held in San Francisco, California. Intel saw great potential in the invention and released the first commercial NOR flash chip in 1988.

Characteristics

Some devices with flash memory can reach speeds of up to 100 MB/s. In general, flash cards have a wide range of speeds and are usually marked at the speeds of a standard CD drive (150 Kb/s). So the indicated speed of 100x means 100 × 150 Kb/s = 15,000 Kb/s= 14.65 Mb/s.

Basically, the volume of a flash memory chip is measured from kilobytes to several gigabytes.

Basically, the volume of a flash memory chip is measured from kilobytes to several gigabytes.

In 2005, Toshiba and SanDisk introduced 1GB NAND chips using multi-level cell technology, where a single transistor can store multiple bits using varying levels of electrical charge on a floating gate.

In September 2006, Samsung introduced an 8 GB chip made using a 40 nm process technology. At the end of 2007, Samsung announced the creation of the world's first MLC (multi-level cell) NAND flash memory chip, made using a 30 nm process technology. The chip capacity is also 8 GB. Memory chips are expected to go into mass production in 2009.

To increase the volume, devices often use an array of several chips. Mostly as of mid-2007 USB devices and memory cards have a capacity from 512 MB to 15 GB. Most large volume USB devices are 128 GB.

NAND flash memory was announced by Toshiba in 1989 at the International Solid-State Circuits Conference. It had a faster write speed and a smaller chip area.

The standardization of NAND flash memory chips is carried out by the Open NAND Flash Interface Working Group (ONFI). The current standard is the ONFI specification version 1.0, released on December 28, 2006. The ONFI group is supported by the largest NAND chip manufacturers: Intel, Micron Technology and Sony.

Operating principle

maximum possible data volumes for crystals using single-bit (SLC) or double-bit (MLC)

The basic data storage cell of flash memory is a floating gate transistor. The peculiarity of such a transistor is that it can hold electrons (charge). It is on this basis that the main types of NAND and NOR flash memory were developed. There is no competition between them, because each type has its own advantages and disadvantages. By the way, hybrid versions such as DiNOR and superAND are built on their basis. In flash memory, manufacturers use two types of memory cells MLC And SLC. .

• Flash memory with MLC (Multi-level cell - multi-level memory cells) cells are more capacious and cheaper, but they are big time access and fewer write/erase cycles (about 10,000).
• Flash memory that contains SLC (Single-level cell) cells has a maximum number of write/erase cycles (100,000) and has shorter access times. Charge reversal (writing/erasing) is accomplished by applying a high potential between gate and source so that the voltage electric field in a thin dielectric between the transistor channel and the pocket was sufficient for the tunneling effect to occur. To enhance the effect of electron tunneling into the pocket during writing, a slight acceleration of the electrons is applied by passing a current through the field-effect transistor channel.

The operating principle of flash memory is based on changing and recording the electrical charge in an isolated area ("pocket") of the semiconductor structure. Reading is performed by a field-effect transistor, for which the pocket acts as a gate. The floating gate potential changes threshold characteristics transistor, which is recorded by the reading circuits. This design is equipped with elements that allow it to work in a large array of the same cells.

NOR and NAND

Layout of six NOR flash cells

Structure of a single column NAND flash with 8 cells

Flash memory differs in the method of connecting cells into an array.

The NOR design uses a classic two-dimensional conductor matrix, in which one cell is installed at the intersection of rows and columns. In this case, the row conductor was connected to the drain of the transistor, and the column conductor was connected to the second gate. The source was connected to a common substrate for all.

NAND design is a three-dimensional array. The matrix is ​​the same as in NOR, but instead of one transistor at each intersection, a column of cells connected in series is installed. This design produces many gate chains at one intersection. The layout density can be increased dramatically (after all, only one gate conductor fits one cell in a column), but the algorithm for accessing cells for reading and writing becomes noticeably more complicated. There are also two MOSFETs installed in each line. Bit line select transistor located between the column of cells and the bit line. And a control ground transistor located in front of the ground (ground select transistor).

NOR technology allows quick access to each cell individually, but the cell area is large. In contrast, NANDs have a small cell area but relatively long access to a large group of cells at once. Accordingly, the scope of application differs: NOR is used as direct program memory of microprocessors and for storing small auxiliary data.

The names NOR and NAND come from the association of the circuitry for including cells in an array with the circuitry of CMOS logic chips.

NAND is most often used for USB flash drives, memory cards, SSD. NOR, in turn, is used in embedded systems.

There were other options for combining cells into an array, but they did not take root.

NOR

NOR Flash memory

The NOR architecture got its name thanks to the logical operation OR - NOT (translated from English NOR). The principle of the logical operation NOR is that over several operands (data, argument of the operation...) it gives a unit value when all operands are equal to zero, and a zero value in all other operations. In our case, operands mean the value of memory cells, which means that in this architecture, a single value on the bit line will be observed only if the value of all cells connected to the bit line is equal to zero (all transistors are closed). This architecture has good random access to memory, but the process of writing and erasing data is relatively slow. The hot electron injection method is used in the writing and erasing process. In addition, the flash memory chip has a NOR architecture and its cell size is large, so this memory does not scale well. Flash memory with the NOR architecture is usually used in storage devices program code. These could be phones, PDAs, System BIOS board... The use of NOR flash, non-volatile memory devices of a relatively small volume, requiring fast access to random addresses and with a guarantee of the absence of faulty elements:

• Embedded program memory for single-chip microcontrollers. Typical volumes range from 1 kbyte to 1 MB.
• Standard random access ROM chips for operation with a microprocessor.
• Specialized chips bootstrap computers (POST and BIOS), DSP processors and programmable logic. Typical volumes are units and tens of megabytes.
• Medium-sized data storage chips, such as DataFlash. Typically equipped with an SPI interface and packaged in miniature cases. Typical volumes range from hundreds of kbytes to the technological maximum.

The maximum capacity of NOR chips is up to 256 MB.

NAND

NAND Flash memory

This type of memory was developed by Toshiba. Due to their architecture, these chips are used in small drives called NAND (logical NAND operation). When executed, the NAND operation produces a value of zero only when all operands are zero, and a value of one in all other cases. As was written earlier, the zero value is the open state of the transistor. As a consequence, the NAND architecture assumes that a bitline has a value of zero when all transistors connected to it are open, and a value of one when at least one of the transistors is closed. Such an architecture can be built by connecting transistors with a bit line not one at a time (as built in the NOR architecture), but in serial series (a column of sequentially connected cells).

This architecture is highly scalable compared to NOR because it allows transistors to be compactly placed on the circuit. In addition, the NAND architecture writes by Fowler-Nordheim tunneling, and this makes it possible to implement quick recording than in the NOR structure. To increase read speed, NAND chips have an internal cache built into them. Like hard disk clusters, NAND cells are grouped into small blocks. For this reason, when reading or writing sequentially, NAND will have a speed advantage. But on the other hand, NAND loses greatly in random access operations and does not have the ability to work directly with bytes of information. In a situation where only a few bits need to be changed, the system is forced to rewrite the entire block, and this, taking into account the limited number of write cycles, leads to a lot of wear on the memory cells.B lately There are rumors that Unity Semiconductor is developing a new generation of flash memory that will be built on CMOx technology. It is assumed that the new memory will replace NAND flash memory and overcome its limitations, which in NAND memory are determined by the architecture of transistor structures. The advantages of CMOx include higher recording density and speed, as well as a more attractive cost. Application areas for the new memory include SSDs and mobile devices. Well, time will tell whether this is true or not.

Record

To write, charges must enter the floating gate, but it is insulated by an oxide layer. The tunneling effect can be used to transfer charges. To discharge, it is necessary to apply a large positive charge to the control gate: the negative charge will leave the floating gate using the tunnel effect. Conversely, to charge a floating gate, a large negative charge must be applied.

Recording can also be achieved using hot media injection. When current flows between source and drain high voltage electrons can travel through the oxide layer and remain in the floating gate. In this case, it is necessary that there is a positive charge on the control gate, which would create a potential for injection.

To MLC for recording different meanings Different voltages and feed times are used.

Each write causes a small amount of damage to the oxide layer, so the number of writes is limited.

Writing to NOR and NAND layouts consists of two stages: first, all transistors in the line are set to 1 (no charge), then the desired cells are set to 0.

At the first stage, the cells are cleaned using a tunnel effect: high voltage is applied to all control gates. Hot carrier injection is used to set a specific cell to 0. The bit line is fed high voltage. Second an important condition This effect is the presence of positive charges on the control gate. Positive voltage is applied only to some transistors; the remaining transistors are supplied with negative voltage. Thus, zero is written only to the cells of interest to us.

• NAND

The first stage in NAND is similar to NOR. A tunnel effect is used to set a cell to zero, as opposed to NOR. A large negative voltage is applied to the control gates of interest to us.

Technological scaling

Due to its highly regular structure and high demand for large volumes, the manufacturing process NAND flash memory decreases more quickly than for less regular DRAM and almost irregular logic (ASIC). High competition between several leading manufacturers only accelerates this process. In the version of Moore's law for logic chips, the number of transistors per unit area doubles in three years, while NAND flash showed a doubling in two years. In 2012, the 19 nm process technology was mastered by a joint venture between Toshiba and SanDisk. In November 2012, Samsung also began production using the 19 nm process technology (actively using the phrase “10 nm class” in marketing materials, denoting some process in the 10-19 nm range).

Reducing the technical process made it possible to quickly increase the volume of NAND flash memory chips. In 2000, flash memory using 180 nm technology had a data volume of 512 Mbit per chip, in 2005 - 2 Gbit at 90 nm. Then there was a transition to MLC, and in 2008 the chips had a capacity of 8 Gbit (65 nm). As of 2010, about 35%-25% of chips had a size of 16 Gbit, 55% - 32 Gbit. In 2012-2014, 64 Gbit crystals were widely used in new products, and the introduction of 128 Gbit modules began (10% at the beginning of 2014), manufactured using 24-19 nm technical processes.

As process technology decreases and approaches the physical limits of current manufacturing technologies, particularly photolithography, further increases in data density can be achieved by moving to more bits per cell (for example, switching from 2-bit MLC to 3-bit TLC), replacing FG cell technology with CTF technology, or switching to a three-dimensional layout of cells on a wafer (3D NAND, V-NAND; however, this increases the process step) . For example, around 2011-2012, all manufacturers introduced air gaps between control lines, which made it possible to continue scaling further than 24-26 nm, and Samsung began mass production of 24- and 32-layer 3D NAND based on CTF technology in 2013-2014. including the version with 3-bit (TLC) cells. The decrease in wear resistance (erasure life) that appears with a decrease in the technical process, as well as an increase in the rate of bit errors, required the use of more complex error correction mechanisms and a reduction in guaranteed recording volumes and warranty periods. However, despite the measures taken, it is likely that further scaling of NAND memory will not be economically justified or physically impossible. Many are being explored possible replacements flash memory technologies, in particular FeRAM, MRAM, PMC, PCM, ReRAM, etc.

3D NAND

NAND circuitry turned out to be convenient for constructing a vertical layout of a block of cells on a chip. Conducting and insulating layers are deposited onto the crystal layer by layer, which form the gate conductors and the gates themselves. Then, many holes are formed in these layers throughout the entire depth of the layers. The structure of field-effect transistors - insulators and floating gates - is applied to the walls of the holes. In this way, a column of ring-shaped field-effect transistors with floating gates is formed.

This vertical structure turned out to be very successful and provided a qualitative leap in flash memory density. Some companies are promoting the technology under their own trademarks, for example V-NAND, BiCS. As of 2016, the number of layers of top products reached 64. .

Data retention period

The pocket insulation is not ideal and the charge gradually changes. The charge storage life declared by most manufacturers for household products does not exceed 10-20 years, although the warranty on media is given for no more than 5 years. At the same time MLC memory has shorter terms than SLC.

Specific environmental conditions, such as elevated temperatures or radiation exposure (gamma radiation and high-energy particles), can dramatically reduce the storage life of data.

With modern NAND chips, when reading, data on adjacent pages within the block may be damaged. Performing a large number (hundreds of thousands or more) of read operations without rewriting can speed up the occurrence of an error.

According to Dell, the length of time an unpowered SSD retains data is highly dependent on the number of write cycles (P/E) and flash memory type, and can range from 3-6 months in worst cases.

Hierarchical structure

Erasing, writing and reading flash memory always occurs in relatively large blocks different sizes, while the size of the erase block is always larger than the write block, and the size of the write block is not less than the size of the read block. Actually, this is a characteristic distinguishing feature of flash memory in relation to classical EEPROM memory.

As a result, all flash memory chips have a clearly defined hierarchical structure. Memory is divided into blocks, blocks consist of sectors, sectors of pages. Depending on the purpose of a particular chip, the depth of the hierarchy and the size of the elements may vary.

For example, a NAND chip can have an erase block size of hundreds of kilobytes, and a write and read page size of 4 kilobytes. For NOR chips, the size of the erase block varies from units to hundreds of kbytes, the size of the write sector - up to hundreds of bytes, the read page size - from units to tens of bytes.

It's no secret that in the modern world, one of the most relevant goods is information. And it, like any other product, must be stored and transferred. Portable storage devices were created for this purpose. In the recent past, this role was played by floppy disks and CDs, capable of storing a very small amount of information in large dimensions. With the development of computer technology, storage media gradually decreased in size, but the volume of data stored in them increased many times over. This led to the emergence of a new portable storage device - the USB flash drive.

Flash memory- a special type of non-volatile, rewritable semiconductor memory.

Let's take a closer look: non-volatile - which does not require additional energy to store data (energy is required only for recording), rewritable - allowing the data stored in it to be changed (rewritten) and semiconductor (solid-state), that is, not containing mechanically moving parts (like regular hard drives or CDs) ), built on the basis of integrated circuits (IC-Chip).

Literally before our eyes, flash memory has transformed from an exotic and expensive means of data storage into one of the most popular storage media. Solid-state memory of this type is widely used in portable players and pocket computers, in cameras and miniature flash drives. The first production samples worked at low speed, but today the speed of reading and writing data to flash memory allows you to watch a full-length movie stored in a miniature chip or run a “heavy” Windows XP-class operating system.

Due to its low power consumption, compact size, durability and relatively high performance, flash memory is ideal for use as storage in portable devices such as digital photo and video cameras, cell phones, laptop computers, MP3 players, digital voice recorders, and etc.

Story

Originally solid state hard drive was developed for high-speed servers and was used for military purposes, but as is usually the case, over time they began to be used for civilian computers and servers.

Two classes of devices arose: in one case, erasure circuits were sacrificed to gain memory high density, and in another case they made a fully functional device with a much smaller capacity.

Accordingly, the efforts of engineers were aimed at solving the problem of the density of the erase circuits. They were crowned with success by the invention of Toshiba engineer Fujio Masuoka in 1984. Fujio presented his development at the International Electron Devices Meeting in San Francisco, California. Intel was interested in this invention and four years later in 1988 it released the first commercial NOR-type flash processor. NAND flash memory architecture was announced a year later by Toshiba in 1989 at the International Solid-State Circuits Conference. The NAND chip had a faster write speed and a smaller circuit area.

It is sometimes argued that the name Flash in relation to the type of memory is translated as “flash”. In fact, this is not entirely true. One version of its appearance says that for the first time in 1989-90, Toshiba used the word Flash in the context of “fast, instant” when describing its new chips. In general, Intel is considered the inventor, introducing flash memory with NOR architecture in 1988.

The advantages of USB flash cards over other drives are obvious:

small dimensions,

very light weight,

quiet operation,

possibility of rewriting,

good resistance to mechanical stress, unlike CDs and floppy disks (5-10 times higher than the maximum permissible for conventional hard drives),

withstands severe temperature changes,

no moving parts, which reduces energy consumption to a minimum,

no connection problems – USB outputs found on almost any computer

large amount of memory,

recording information into memory cells,

Information storage period is up to 100 years.

Flash memory consumes significantly (about 10-20 times or more) less energy during operation.

It should also be noted that to work with a USB flash drive, you do not need any third-party programs, adapters, etc. The device is recognized automatically.

If you write to a flash drive 10 times a day, it will last for about 30 years.

Operating principle

The operating principle of semiconductor flash memory technology is based on changing and recording the electrical charge in an isolated area (pocket) of the semiconductor structure.

The change in charge (“write” and “erase”) is accomplished by applying a high potential between the gate and source so that the electric field strength in the thin dielectric between the transistor channel and the pocket is sufficient to cause a tunneling effect. To enhance the effect of electron tunneling into the pocket during writing, a slight acceleration of the electrons is applied by passing a current through the field-effect transistor channel.

Schematic representation of a floating gate transistor.

Between the control gate and the channel through which current flows from source to drain, we place the same floating gate, surrounded by a thin layer of dielectric. As a result, when current flows through such a “modified” field-effect transistor, some high-energy electrons tunnel through the dielectric and end up inside the floating gate. It is clear that while the electrons were tunneling and wandering inside this gate, they lost some of their energy and practically cannot return back. SLC and MLC devices

There are devices in which the elementary cell stores one bit of information and several. In single-bit cells, only two charge levels are distinguished on the floating gate. Such cells are called single-level cells. single-level cell SLC). In multi-bit cells, more charge levels are distinguished; they are called multi-level. multi-level cell, MLC). MLC devices are cheaper and more capacious than SLC devices, but the access time and number of rewrites are worse.

Audio memory

A natural development of the idea of ​​MLC cells was the idea of ​​recording an analog signal into the cell. Such analog flash chips are most widely used in sound reproduction. Such microcircuits are widely used in all kinds of toys, sound cards, etc.

Nor flash memory

Design NOR uses a classic two-dimensional matrix of conductors (“rows” and “columns”) in which one cell is installed at the intersection. In this case, the conductor of the rows was connected to the drain of the transistor, and the conductor of the columns to the second gate. The source was connected to a common substrate for all. With this design, it was easy to read the state of a particular transistor by applying a positive voltage to one column and one row.

At the core of this type Flash memory uses the OR-NOR algorithm, since in a transistor with a floating gate, too low a gate voltage means one. This type of transistor consists of two gates: floating and control. The first gate is completely insulated and has the ability to retain electrons for up to ten years. The cell also consists of a drain and a source. When voltage is applied to the control gate, an electric field is generated and the so-called tunnel effect occurs. Most of the electrons are transferred (tunneled) through the insulator layer and enter the floating gate. The charge on the floating gate of the transistor changes the drain-source "width" and conductivity of the channel, which is used for reading. Writing and reading cells are very different in power consumption: for example, flash drives consume more current when writing than when reading (consuming very little power). To delete (erase) data, a sufficiently high negative voltage is applied to the control gate, which leads to reverse effect(electrons from the floating gate move to the source using the tunnel effect). In the NOR architecture, there is a need to connect a contact to each transistor, which greatly increases the size of the processor. This problem is solved using the new NAND architecture.

What is Flash Memory?

Flash Memory/USB drive or flash memory is a miniature storage device used as an additional storage medium for information. The device connects to a computer or other reading device via a USB interface.

A USB flash drive is designed to be read repeatedly over a specified lifespan, which typically ranges from 10 to 100 years. You can write to flash memory a limited number of times (about a million cycles).

Flash memory is considered more reliable and compact compared to hard drives(HDD) because it has no moving mechanical parts. This device is quite widely used in the production of digital portable devices: photo and video cameras, voice recorders and MP3 players, PDAs and mobile phones. Along with this, Flash Memory is used to store firmware in various equipment, such as modems, PBXs, scanners, printers or routers. Perhaps the only drawback of modern USB drives is their relatively small volume.

History of Flash Memory

The first flash memory appeared in 1984, it was invented by Toshiba engineer Fujio Masuoka, whose colleague Shoji Ariizumi compared the principle of operation of this device with a photo flash and first called it “flash”. The public presentation of Flash Memory took place in 1984 at the International Electronic Devices Seminar held in San Francisco, California, where Intel became interested in this invention. Four years later, its specialists released the first commercial flash processor. The largest manufacturers of flash drives at the end of 2010 were Samsung, occupying 32% of this market, and Toshiba - 17%.

How does a USB drive work?

All information written to a Flash drive and stored in its array, which consists of floating gate transistors called cells. IN conventional devices with single-level cells, each of them “remembers” only one bit of data. However, some new chips with multi-level cells (multi-level cell or triple-level cell) are capable of storing a larger amount of information. In this case, a different electrical charge must be used on the floating gate of the transistor.

Key Features of a USB Drive

The capacity of currently available flash drives ranges from several kilobytes to hundreds of gigabytes.

In 2005, specialists from Toshiba and SanDisk presented a NAND processor, the total volume of which was 1 GB. When creating this device, they used multi-level cell technology, where a transistor is capable of storing several bits of data using a different electrical charge on a floating gate.

In September of the following year, Samsung presented to the public a 4-gigabyte chip developed on the basis of a 40-nm technological process, and at the end of 2009, Toshiba technologists announced the creation of a 64 GB flash drive, which was launched into mass production at the beginning of next year. year.

In the summer of 2010, the presentation of the first 128 GB USB drive in the history of mankind, consisting of sixteen 8 GB modules, took place.

In April 2011, Intel and Micron announced the creation of an 8 GB MLC NAND flash chip with an area of ​​118 mm, almost half the size of similar devices, the mass production of which started at the end of 2011.

Types of memory cards and flash drives

It is used mainly in professional video and photo equipment, since it has rather large dimensions (43x36x3.3 mm), as a result of which it is quite problematic to install a Compact Flash slot in mobile phones or MP3 players. At the same time, the card is considered not very reliable, and also does not have high speed data processing. The maximum allowable capacity of Compact Flash currently reaches 128 GB, and the data copying speed has increased to 120 MB/s.

RS-MMC/Reduced Size Multimedia Card- a memory card that is half the length of a standard MMC card - 24x18x1.4 mm and weighs about 6 grams. At the same time, all other characteristics and parameters of a regular MMC card are preserved. To use RS-MMC cards, you must use an adapter.

MMCmicro- a miniature memory card with dimensions of only 14x12x1.1 mm and designed for mobile devices. To use it, you must use a standard MMC slot and a special adapter.

Despite the parameters and dimensions of 32x24x2.1 mm being very similar to the MMS card, this map cannot be used with a standard MMC slot.

SDHC/SD High Capacity is a high capacity SD memory card, known modern users as SD 1.0, SD 1.1 and SD 2.0 (SDHC). These devices differ in the maximum amount of data that can be stored on them. Thus, there are capacity restrictions in the form of 4 GB for SD and 32 GB for SDHC. However, the SDHC card is backward compatible with SD. Both options come in three physical size formats: standard, mini and micro.

microSD/Micro Secure Digital Card- this is the most compact according to 2011 data removable devices flash memory, its dimensions are 11x15x1 mm, which allows it to be used mobile phones, communicators, etc. The write protection switch is located on the microSD-SD adapter, and the maximum possible card capacity is 32 GB.

Memory Stick Micro/M2- a memory card whose format competes in size with microSD, but the advantage remains with Sony devices.