Is the 64 GB storage space on your Android smartphone still not enough to hold your data? California tech startup Crossbar might just a solution to this as they announced a new kind of memory chip that will soon be replacing the existing NAND type of chips. The Resistive RAM or ReRAM can store up to 1TB of data and is 20 times faster than the best memory chips in the market today. It’s also half the size of current NAND chips and uses 20 times less power while lasting 10 times longer.
Crossbar announced that “This new generation of non-volatile memory will be capable of storing up to 1TB of data on a single 200mm2 chip, enabling massive amounts of information, such as 250 hours of HD movies, to be stored and played back from an IC smaller than a postage stamp.”
The company announced some key benefits which include
- Highest Capacity: Up to 1 Terabyte (TB) of Storage on a Single Chip; Multiple Terabytes with 3D Stacking
- Lowest Power: Extends Battery Life to Weeks, Months or Years
- Highest Performance: 20x Faster Write than NAND
- Easiest SOC Integration: Simple Stacking on Logic in Standard CMOS at Most Advanced Nodes
- Most Reliable: 10x the Endurance of NAND; Approaching DRAM Reliability
To explain this technology further, RERAM works by creating resistance instead of storing electrical charge. So when an electric current is applied it changes the resistive state of the material which can be translated to either a 1 or a 0. Most of the research done on this technology was in finding the right material to use and measuring the resistive state.
Crossbar says that they are ready to mass produce this product but will initially target low-density applications such as microcontrollers. The company may be small right now but they are aiming to demonstrate the capability of their product first to draw in more investors. Right now major companies such as SK Hynix, Panasonic, and HP are said to be working on ReRAM designs.
One particular interest is applying this technology to smartphones and tablets. This allows consumers to not only store vast amounts of data but access them quickly as well.