What is ReRAM or Resistive Random Access Memory

Do you want to know what is ReRam memory and how does it work. You are on the right spot to know the answer of this question.

ReRAM is a type of memory that combines the benefits of NAND Flash memory and DRAM memory. This resistive memory was proposed years ago, and the first to announce its production was the manufacturer Crossbar, when it announced the first chips in 2013.

The idea was to create memory devices capable of being up to 100 times faster than conventional RAM at that time.

Despite that, this type of resistive RAM has not yet dominated the market and has become the de facto standard, but remains somewhat marginal.

See Also: What is RAM Memory

In this post you will understand why, as well as learn everything you need to know about this new concept of memory cell. But let’s check below first what is NAND flash and DRAM memory?

What is NAND Flash Memory?

NAND flash memory is a type of flash storage memory that stores data in arrays of memory cells that are made using floating gate transistors.

NAND flash memory is a form of non-volatile storage, which means it keeps your data even when the power is out.

NAND memory is non-volatile, which means that it retains the stored data even if it is powered off.

Flash storage technology is a chip-based non-volatile store unlike DRAM, which does not require a constant source of electricity.

NAND flash memory is the most common data storage format in solid-state drives (SSDs), integrated memory cards, and USB devices.

In this respect, NAND flash memory is similar to other secondary data storage devices, such as hard drives and optical media.

Therefore, it is well suited for use in large-scale data storage devices, such as memory cards, memory and solid-state drives (SSDs).

An important goal in the development of NAND flash memory has been to reduce the cost per bit and increase the maximum capacity per chip, so that flash memory can compete with magnetic storage devices such as hard drives.

Manufacturers made advances in NAND flash memory density, performance, and reliability throughout the 2000s and 2010s.

As Taking advantage of new cell design technologies, such as multi-level cells (MLCs) that offer two bits per cell.

Triple level cells (TLC) store three bits per cell and quad level cells (QLC) store four bits per cell. Shrinking cells using the aforementioned cell-level technologies causes cell-to-cell interference, which reduces data integrity in NAND flash memory due to the reduction of the matrix surface caused by the smaller cells.

See Also: Is 16 GB RAM Enough for your Gaming PC

The reads make NAND flash more of a data storage device than RAM- type memory, since it does not allow random access to data, which is what many applications require.

This feature makes NAND flash especially attractive to industrial users who need a form of data storage that is robust enough to safeguard data even in the event of a power loss.

NAND flash has the advantages of large capacity and high writing speed, suitable for storing large amounts of data, so it has been increasingly used by the industry.

For example, in embedded products such as digital cameras, compact USB flash drives and others. MP3 players, digital cameras, and USB sticks use NAND technology.

The NAND type is mostly found in memory cards, USB flash drives, solid-state drives (those made after 2009), landline phones, Smartphones, and other similar products, which are used to store and share data.

In short, NAND Flash is a type of Non-Volatile Memory (NVM) technology that stores data in arrays of memory cells, which are fabricated using either charge capture technology or MOSFET floating gate transistors.

Vertical NAND memories (V-NAND) or 3D NAND they stack memory cells vertically and use Charge Trap flash architecture.

The two on the left are planar NAND, but the structure of the memory cells is different, going from floating gate to charge trap flash, i.e. 2D CTF (Charge Trap Flash) shown above.

What is DRAM Memory?

Dynamic Random Access Memory (DRAM) is a type of RAM for computers that has been in use since the mid-1960s.

DRAM stands for Dynamic Random Access Memory, and is the type of semiconductor memory seen in RAM and GPUs (also known as graphics cards). Dynamic RAM (DRAM) is a type of semiconductor memory that uses capacitors to store bits.

See Also: How to change RAM memory

DRAM is volatile memory, meaning data in a memory is lost when power is removed. All types of RAM, including DRAM, are volatile memory, storing data bits on transistors.

Dynamic Random Access Memory (DRAM or Dynamic RAM) is a type of semiconductor random access memory that stores each bit of data in a memory cell.

These cells are usually made up of a small capacitor and a transistor and generally based on the metal oxide semiconductor (MOS) technology.

DRAM is a common type of random access memory (RAM) used in personal computers (PCs), workstations, and servers.

DRAM can store a larger amount of data than SRAM, and because it is considerably cheaper to manufacture.

DRAM is the most common type of memory in computer systems as the primary storage medium, leaving flash for secondary and the SRAM for memories of smaller capacity, but faster.

Consequently, SRAM (pronounced ES-RAM) is faster, but it is also more expensive, making DRAM the most common memory in computer systems.

In general, RAM is much faster than other types of memory used by your computer, and DRAM is even faster.

With more RAM in a computer, there is less need for the CPU to read data from secondary memory (with higher latencies), allowing the computer to run faster.

Once the operating system is loaded, the computer uses RAM memory to temporarily maintain software processes while the Central Processing Unit (CPU) performs other tasks.

One of the largest applications of DRAM is main storage (called RAM) in modern computers and graphics cards (where main RAM is called graphics storage).

The use of DRAM as the main memory of the computer is one of its most common applications. DRAM is a semiconductor memory option that a system designer can choose to build a computer.

Dynamic Random Access Memory (DRAM) is a type of semiconductor memory that is typically used for data or programming code required to run a computer’s processor.

See Also: What does RAM Memory do

Despite requiring additional refresh circuitry to be designed into the computer’s processor bus, DRAM was the most widely used type of memory until recently, as it was much cheaper to produce than non-volatile static RAM.

What is ReRAM Memory?

ReRAM memory or resistive RAM memory is a type of non-volatile memory. If it is disconnected from the power source it will continue to maintain the data, unlike what would happen with conventional volatile RAM memories.

Volatile RAM memories need to be constantly connected to electricity so as not to be erased. In addition, as it is still a fast random access memory.

It also has the advantages of DRAM, so it could technically be considered a hybrid between DRAM and NAND flash.

To achieve this, ReRAM memory cells are made up of 3 layers. The upper one is the metallic electrode, the lower one is the other electrode, in this case also metallic.

The central one is a type of dielectric (a solid-state material known as memristence) that will act as a switch, being able to store bits (ones and zeros).

Its way of working would be similar to a PCM or CBRAM, without transistors that store charge or anything similar.

A ReRAM memory cell works thanks to the transfer of metallic ions from the metallic electrode to the non-metallic electrode, creating a conductive filament between them when the dielectric allows it.

In this way, if they are not connected it will be a value 0 (OFF), while if they are connected it will be a value 1 (ON).

That not only makes it fast and simple, but also makes the memory controller much simpler and more reliable, as well as cheaper to manufacture.

See Also: What is MT/s ram

On the other hand, this type of memory is up to 20 times more efficient than a conventional NAND, and can last up to 10 times more cycles than NAND, which makes them much more interesting.

Disadvantages of  ReRAM

It seems that ReRAM or RRAM are all advantages. This does not square with the fact of not seeing that this memory has become the predominant one in the market.

And it is not all advantages, it also has some drawbacks that must be analyzed to understand its reduced impact on the market.

One of the most prominent problems with ReRAM is that its profit margins are not as high as other memories that already exist, since it is a technology that has yet to mature.

In addition, there is some uncertainty on the part of manufacturers when it comes to using memristors.

This “magical” component that promised to be the substitute for the conventional transistor and that finally has not been able to be. On the other hand, manufacturers do not agree on how the memsistor can be better manufactured.

Whether with tantalum oxide, or with Ag/a-Si-Si structures similar to those of CBRAM, with perovskites (PCMO), transition metal oxides such as NiO or TiO2, and even phase change chalcogenides, such as Ge2Db2Te5.

Some of these substances or materials, let’s remember, are quite expensive compared to the abundant and cheap silicon.

Conclusion

Despite all the advances and advantages of RRAM, this type of memory is not intended to replace current flash drives, at least in the short-medium term.

And of course you won’t do it with conventional RAM either. So, the current concepts will be the ones that accompany us for several more years as sometimes it’s not just about performance.

See Also: How to Repair RAM Memory

The industry analyzes the market with a magnifying glass, and is especially focused on achieving profitability, the maximum possible.

And this implies staying with technologies that already exist so as not to have to modify all the production processes in a too obvious way.

For the same reason, in the semiconductor industry, alternatives to conventional silicon are being sought, and many of the novelties involve materials that achieve improvements.

But that can continue to be manufactured using the same manufacturing processes that are currently used. And that means limiting the freedom to innovate in some cases.