Resistive RAM (ReRAM) technology has been biding its time in the sun for several years now. Despite its placement on the roadmaps of several chipmakers and as part of new architectures like “The Machine” project based on HPE shutter memristors, it never took off as a breakthrough memory technology, even. so the concept makes perfect sense.

Manufacturing capacity, cost, and other factors played a role, but for companies like Crossbar, which has banked an entire company on the emerging role of ReRAM, the gamble has yet to pay off. The same company today announced a possible new life for ReRAM technology, in terms of security.

CrossBar, Inc. claims that ReRAM technology can be used in the creation of non-clonable physical cryptographic keys (PUF) because each key is specific to the individual integrated circuit.

This idea has some support, notably from Dr. Bertrand Cambou of the Nanotechnology and Cybersecurity group at North Arizona University. He says that after evaluating other PUF approaches, ReRAM has some advantages. “Due to its unique stochastic and electrical characteristics, CrossBar’s ReRAM PUF enables more secure systems than existing PUF technologies. Cambou discusses these other approaches in relation to ReRAM here in detail. This is a useful description of how PUFs suit ReRAM as well.

“The approach to generate PUFs on ReRAM arrays shows promise based on statistical analysis of resistive devices,” Cambou said. He adds: “Resistive RAM is an attractive memory technology for designing secure applications, PUFs and RNGs. It is energy efficient, fast, compact, and less susceptible to side channel attacks than flash memory.

To be fair, researching the role of ReRAM technology in PUFs and security is not a new idea. One of the first appearances of the concept came from Panasonic’s semiconductor group in 2016 using a 40nm ReRAM test chip to test the idea (with moderate success). Despite 2016 research so far, it’s still a bit marginal in terms of what use cases we might find and, like ReRAM itself, has all the right components but not the right ‘IT factor’ to break through. a path in the wider market. .

It must be very frustrating for CrossBar. The company finds itself in another tough market after ReRAM failed to materialize as the next big memory innovation. There are a lot of tech already on the PUF box (all of which use SRAM, by the way, which is cheap). Compared to SRAM-based PUF approaches, their method allows for a higher degree of randomness, lower bit error rates, and all this “without the need for fuzzy extractors, auxiliary data, or correction codes. heavy mistake, ”according to Mark Davis, president of CrossBar. .

In terms of ReRAM more broadly, it’s worth taking a closer look (we’re looking for one) on why ReRAM never took off. It can scale below 10nm, has exceptional read latencies, and can write a little faster than NAND. It doesn’t require any super sophisticated crafting magic. So why has the market switched to ReRAM? Dive deeper on this site.

CrossBar says its business is now available in multiple CMO foundry process nodes to achieve even higher densities and more tightly integrated devices. By the way, the company stays afloat by licensing SoC and memory companies as out-of-the-box and custom IP cores. It was first released in 2010, at the start of next-gen memory, and while the angle of security applications may help them continue to fight ReRAM, it’s hard to say what it will look like. future of technology in the years to come.

Subscribe to our newsletter

Featuring the week’s highlights, analysis, and stories straight from us to your inbox with nothing in between.
Subscribe now


Source link