06-05-2021 | | By Robin Mitchell
Recently, Weebit Nano Ltd filed a patent describing new technologies in the field of ReRAM. So what is ReRAM, what does the patent describe, and how can ReRAM help with memory technologies of the future?
Resistive Random-Access-Memory, or ReRAM, is a memory technology that uses electrical resistance to store information. However, the resistive memory used can change its resistance depending on how much current has flowed through it, and thus this allows the memory to be reprogrammable.
ReRAM, unlike DRAM, is non-volatile meaning that it remembers the contents of its memory even when power is removed. Furthermore, the ability to retain its data also removes the need for refresh cycles, reducing the complexity of designs when using ReRAM with processors.
Most ReRAM designs utilise a special phase-changing material (such as HfO2) controlled using a standard NMOS transistor. When a high current is passed through the resistive material, its resistance changes and the material's resistance is also dependent on the direction of the current. Therefore, when the bit is read, a small gate voltage is applied to the transistor which allows current to flow through the phase-changing material, and the current that results determine the state of the bit.
Weebit Nano is an Israeli company formed in 2015 to develop ReRAM for next-generation computing systems. The growing need for non-volatile memory in computing systems that allows for unlimited read and write cycles presents ReRAM a real opportunity to become an established memory technology.
Unlike most other ReRAM technologies, Weebit Nano has focused on using Silicon Oxide as their phase-changing material. Their memory cells consist of metal contacts (Titanium), on either side of their phase-changing structure. Their technology allows for up to 1 million read / write cycles and has demonstrated a working 1Mbit device using 40nm technology.
Recently, Weebit Nano filed two patents surround the development of its ReRAM technology. The first patent filed by Weebit Nano describes circuitry that will aid in increasing the performance characteristics of their technology (mainly addressing access time and power consumption). The second patent filed by Weebit Nano describes how ReRAM can be expanded in used to provide double memory speed access by mainstream computing systems.
With the patents filed, Weebit Nano is also looking at integrating their memory into a RISC-V microprocessor to demonstrate how it can be used in real-world applications. However, the commercial market will not be able to expect ReRAM devices until the end of 2023 while the company raises funds and further develops its technology.
One of the biggest advantages ReRAM has over FLASH is its high endurance cycles. This allows it to be more reliably used in applications that need frequent bidirectional memory access (i.e. read and write). However, even at 1 million cycles, ReRAM would not be able to replace memory technologies such as SRAM and DRAM.
If ReRAM could be designed to have an unlimited number of read/write cycles, it could easily replace multiple memory technologies to become the first truly universal memory. However, current FLASH memory technology has the advantage of 3D stacking which gives FLASH a major memory density advantage over ReRAM, and SRAM can be made using extremely small transistors.
This does not mean that research into ReRAM should be stopped or hindered. If anything, it is a good reason to explore such memory technologies. For example, future CPUs and MPUs could be integrated with onboard ReRAM that behaves as a frequent local memory storage area that can retain its contents even when powered down. This would be ideal for applications such as servers and data centres where power outages can be disastrous. A ReRAM array could copy the contents of DRAM and restore the contents of memory upon power-up.