FRAM technology
Memory in modern microcontrollers can be divided on the basis of dependence on power supply. DRAM and SRAM technologies belong to volatile memory, to non-volatile memory - EEPROM / Flash This division exists due to the fact that DRAM / SRAM have much better speed in comparison with nonvolatile memory. But what would have happened if there had been a non-volatile memory, which is not inferior to volatile memory in terms of read / write speed and power consumption? It turns out that such technologies exist. One of the representatives of this class of memory technology is FRAM or FeRAM. For details, I ask under the cat.
So, FeRAM or Ferromagnetic Random Access non-volatile Memory is a type of memory whose principle of operation is based on the hysteresis effect in a ferroelectric. When applied to the cell of the electric field, it changes its polarization, moving to another part of the hysteresis loop. Due to this, it is possible to obtain two states that are well distinguishable in energy, and this is sufficient to create a memory based on such a cell. This is well illustrated by gifs from the Fujitsu website - one of the main producers of FRAM.
Fig.1 The principle of operation of FRAM
')
In order to understand what advantages this gives over classical types of memory, it is also necessary to recall the basic principles of the operation of other types of memory.
The principle of operation of DRAM (Dynamic RAM) is based on reading and changing the charge of the capacitor. If the capacitor is charged, the cell is in the “1” state; if it is discharged, it is in the “0” state. Just like an umbrella. To increase the speed in memory cells, small capacitors are used, the charge from which flows relatively quickly. Therefore, to ensure the safety of information information has to be regenerated. DRAM is used as RAM on modern computers due to its low cost (compared to SRAM) and high speed (compared to disk drives).
Fig.2 Typical DRAM cell
SRAM (Static RAM) memory is much more complicated than DRAM, and therefore much more expensive. Its principle of operation is based on the use of CMOS transistors. When you combine several transistors, you can get a trigger - a cell that preserves a certain logical state. For this type of memory, there is no need for regeneration of states, but nevertheless, in the absence of power, data is lost, i.e. memory remains volatile. This kind of memory is faster than DRAM. Since such memory is much more expensive than DRAM, it is used where very little response time is required — in the processor’s cache memory.
Fig.3 Six-transistor SRAM cell
Modern Flash and EEPROM based on the use of transistors with the so-called floating gate. Electrons are injected into the “pocket” of the semiconductor structure, and their presence / absence can be registered externally. This is the property that allows you to use such structures as memory. Though the charge from the pocket leaks, but it happens rather slowly (~ 10-20 years), which allows using EEPROM / Flash as non-volatile memory. Flash is used to store program code in microcontroller devices, as well as in memory cards.
Fig.4 Floating gate transistor
What is FRAM better than these types of memory?
The main advantage of FRAM over SRAM is non-volatility. When the power supply to the memory chip is cut off, it retains its previous state. When this speed of these types of memory is comparable to each other - the FRAM write cycle takes 150 nanoseconds versus 55 nanoseconds in SRAM according to the Fujitsu website . But FRAM has a limited (albeit huge - 10 ^ 13) number of rewriting cycles, whereas SRAM has no such limitations. DRAM strongly loses power consumption FRAM due to the need for data regeneration. Therefore, DRAM is not used in devices that are sensitive to power consumption.
Nevertheless, although FRAM is comparable in performance to SRAM, the main application potential is tied to significant advantages over Flash-memory. First of all, this is a huge speed. From the same link to the Fujitsu website, the time for one write cycle to Flash is about 10 microseconds. Here we should mention the peculiarity of the use of flash-memory - the recording and erasing in it is done in fairly large blocks. Therefore, overwriting one byte in a flash is a very expensive pleasure both in time and in power consumption — you need to save a data block somewhere, change the byte in it, completely erase the corresponding section of the block and overwrite the updated data into it. Here, by the way, one more advantage of FRAM is a memory with random access, which means that it is possible to change individual bits in it without touching neighboring ones. But even when writing large blocks of data, FRAM is an order of magnitude faster. For example, in Texas Instrument controllers, a 13 kB block size recording takes 10 ms in FRAM versus 1 second in Flash ( proof ). Another disadvantage of Flash is the very limited number of rewriting cycles, on the order of 10 ^ 5.
When I found out about all these properties, I still have one question - why has FRAM still not killed Flash? Indeed, all FRAM characteristics are orders of magnitude better than flash characteristics. Here the main disadvantages of ferromagnetic RAM surfaced. First of all, it is low density of information placement due to the peculiarities of the technology. From this drawback follows another - the capacity of FRAM drives cannot be made large enough. Fujitsu offers memory circuits up to 4 Mbit, which can not be compared with multi-gigabyte flash drives. Another drawback is the relatively high cost of memory. Today, FRAM occupies a completely tiny share of the semiconductor device market.
For which applications is the FRAM memory type optimal? FRAM is good enough in microcontrollers in combination with a small amount of SRAM. Actually, this is the very application that attracted me to this type of memory. For example, Texas Instruments has released a line of FRAM microcontrollers with completely missing Flash / EEPROM. The code is written to them in the FRAM segment, and data in the same FRAM can be accessed in the same way as normal RAM memory. Such an application is convenient where there is a significant amount of data that can often be rewritten. For example, a portable logger for which power consumption is important. You can record data in FRAM for a certain time, then analyze and, for example, send data on average values ​​over the wireless channel. Flash memory is inconvenient for such use - it will quickly land the battery, and due to the limited recording cycles, problems with damaged memory cells may appear after a while. Thus, FRAM is beneficial for low-power applications with a relatively large amount and high frequency of writing to non-volatile memory. In general, TI on its website indicates in which areas they think such a memory is most convenient.
I hope I managed to draw your attention to this interesting and unusual technology, about which, unfortunately, there is almost no information on Habré / Hiktimes.
So, FeRAM or Ferromagnetic Random Access non-volatile Memory is a type of memory whose principle of operation is based on the hysteresis effect in a ferroelectric. When applied to the cell of the electric field, it changes its polarization, moving to another part of the hysteresis loop. Due to this, it is possible to obtain two states that are well distinguishable in energy, and this is sufficient to create a memory based on such a cell. This is well illustrated by gifs from the Fujitsu website - one of the main producers of FRAM.
Fig.1 The principle of operation of FRAM
')
In order to understand what advantages this gives over classical types of memory, it is also necessary to recall the basic principles of the operation of other types of memory.
The principle of operation of DRAM (Dynamic RAM) is based on reading and changing the charge of the capacitor. If the capacitor is charged, the cell is in the “1” state; if it is discharged, it is in the “0” state. Just like an umbrella. To increase the speed in memory cells, small capacitors are used, the charge from which flows relatively quickly. Therefore, to ensure the safety of information information has to be regenerated. DRAM is used as RAM on modern computers due to its low cost (compared to SRAM) and high speed (compared to disk drives).
Fig.2 Typical DRAM cell
SRAM (Static RAM) memory is much more complicated than DRAM, and therefore much more expensive. Its principle of operation is based on the use of CMOS transistors. When you combine several transistors, you can get a trigger - a cell that preserves a certain logical state. For this type of memory, there is no need for regeneration of states, but nevertheless, in the absence of power, data is lost, i.e. memory remains volatile. This kind of memory is faster than DRAM. Since such memory is much more expensive than DRAM, it is used where very little response time is required — in the processor’s cache memory.
Fig.3 Six-transistor SRAM cell
Modern Flash and EEPROM based on the use of transistors with the so-called floating gate. Electrons are injected into the “pocket” of the semiconductor structure, and their presence / absence can be registered externally. This is the property that allows you to use such structures as memory. Though the charge from the pocket leaks, but it happens rather slowly (~ 10-20 years), which allows using EEPROM / Flash as non-volatile memory. Flash is used to store program code in microcontroller devices, as well as in memory cards.
Fig.4 Floating gate transistor
What is FRAM better than these types of memory?
The main advantage of FRAM over SRAM is non-volatility. When the power supply to the memory chip is cut off, it retains its previous state. When this speed of these types of memory is comparable to each other - the FRAM write cycle takes 150 nanoseconds versus 55 nanoseconds in SRAM according to the Fujitsu website . But FRAM has a limited (albeit huge - 10 ^ 13) number of rewriting cycles, whereas SRAM has no such limitations. DRAM strongly loses power consumption FRAM due to the need for data regeneration. Therefore, DRAM is not used in devices that are sensitive to power consumption.
Nevertheless, although FRAM is comparable in performance to SRAM, the main application potential is tied to significant advantages over Flash-memory. First of all, this is a huge speed. From the same link to the Fujitsu website, the time for one write cycle to Flash is about 10 microseconds. Here we should mention the peculiarity of the use of flash-memory - the recording and erasing in it is done in fairly large blocks. Therefore, overwriting one byte in a flash is a very expensive pleasure both in time and in power consumption — you need to save a data block somewhere, change the byte in it, completely erase the corresponding section of the block and overwrite the updated data into it. Here, by the way, one more advantage of FRAM is a memory with random access, which means that it is possible to change individual bits in it without touching neighboring ones. But even when writing large blocks of data, FRAM is an order of magnitude faster. For example, in Texas Instrument controllers, a 13 kB block size recording takes 10 ms in FRAM versus 1 second in Flash ( proof ). Another disadvantage of Flash is the very limited number of rewriting cycles, on the order of 10 ^ 5.
When I found out about all these properties, I still have one question - why has FRAM still not killed Flash? Indeed, all FRAM characteristics are orders of magnitude better than flash characteristics. Here the main disadvantages of ferromagnetic RAM surfaced. First of all, it is low density of information placement due to the peculiarities of the technology. From this drawback follows another - the capacity of FRAM drives cannot be made large enough. Fujitsu offers memory circuits up to 4 Mbit, which can not be compared with multi-gigabyte flash drives. Another drawback is the relatively high cost of memory. Today, FRAM occupies a completely tiny share of the semiconductor device market.
For which applications is the FRAM memory type optimal? FRAM is good enough in microcontrollers in combination with a small amount of SRAM. Actually, this is the very application that attracted me to this type of memory. For example, Texas Instruments has released a line of FRAM microcontrollers with completely missing Flash / EEPROM. The code is written to them in the FRAM segment, and data in the same FRAM can be accessed in the same way as normal RAM memory. Such an application is convenient where there is a significant amount of data that can often be rewritten. For example, a portable logger for which power consumption is important. You can record data in FRAM for a certain time, then analyze and, for example, send data on average values ​​over the wireless channel. Flash memory is inconvenient for such use - it will quickly land the battery, and due to the limited recording cycles, problems with damaged memory cells may appear after a while. Thus, FRAM is beneficial for low-power applications with a relatively large amount and high frequency of writing to non-volatile memory. In general, TI on its website indicates in which areas they think such a memory is most convenient.
I hope I managed to draw your attention to this interesting and unusual technology, about which, unfortunately, there is almost no information on Habré / Hiktimes.
Source: https://habr.com/ru/post/390389/
All Articles