The annual “Top 100 Innovators” list has been published, highlighting the world’s most innovative companies. In addition to Amazon, Apple and Google, some may be surprised to see Micron making the list – for the sixth straight year. Many, even in the technology industry, consider memory and storage to be static, and certainly not innovative. Indeed, memory and storage have traditionally been considered commodities, subject to brutal competition based on price.
But a funny thing happened over the years. The number of worldwide memory and storage players has narrowed to a relative few. And a key, if underappreciated characteristic of the surviving companies is innovation.
Without advances in how the world physically stores and retrieves data, today’s most useful devices and algorithms would not exist. Innovations in memory and storage – workmanlike, methodical, critical – are the untold story of the data age
Consider the advent of multi-patterning. In the early 2000’s, lithography tools, which print images of circuits and other components on chips, weren’t able to create the increasingly smaller components postulated by Moore’s Law. As we know, regularly “cramming” more components onto a given area of silicon has been the fundamental technology driver for decades.
Micron was foundational in overcoming the challenge by pioneering pitch double-patterning. This technology uses advanced atomic layer deposition (ALD) and etching techniques to increase the density of a chip far beyond what is otherwise possible. The method is now used not just by Micron but by NAND and DRAM manufacturers worldwide. Without pitch double- and quadruple-patterning, the ability to store increasing amounts of data at lower prices – fundamental to the advent of many technologies – would have been impossible.
Shrinking Components on DRAM Chips
Before that, Micron was the leader in shrinking the size of components on DRAM chips for most of the 1990s. This was enabled by low-mask count process technology made possible by first-to-market process technologies such as chemical vapor deposition (CVD) of Ti, TiN and W films for making high-aspect ratio contact structures, chemical mechanical planarization (CMP) for planarization of topological features as well as high aspect ratio etching (HARC) and atomic layer deposition (ALD) of high K films for DRAM capacitor formation.
These and many other innovations from Micron were key to keeping Moore’s Law alive over the years. In 2006, IEEE Spectrum reviewed thousands of patent portfolios and named Micron’s the best in its “Pipeline Power” ranking, measuring innovations that frequently lead to other innovations.
3D NAND Leadership
Micron led the pack in 3D NAND technology, delivering the industry’s smallest and highest-density 3D memory die. Micron uses a unique architectural approach, placing all the flash memory layers on top of the logic array, maximizing the use of space while enabling large memory capacity. This technique has helped reduce costs-per-bit even as the ability to shrink features on chips, as predicted in the original Moore’s Law, has slowed due to technical barriers.
3D XPoint Technology
Also, Micron helped spearhead development of 3D XPoint™ Technology. This non-volatile memory has much greater density than DRAM but is also exhibits much lower latency and greater endurance than NAND. These features will be a boon for applications that need fast, persistent memory; everything from big data to transactional server workloads.
DNA For Data Storage?
Looking farther into the future, Micron is exploring techniques that might sound like fantasy. For example, researchers currently use DNA to help cure diseases, solve crimes and more. But what if we could use DNA to store data? It turns out that DNA as a storage medium is amazingly dense, durable and energy-efficient. Theoretically, an amount of DNA the size of a sugar crystal could be used store billions of gigabytes of information. Such capabilities don’t just sound cool – they might soon be required. Data volumes are exploding and the amount of memory and storage produced each year is lagging. Micron is collaborating with and supporting research in academia as well as through research consortia to explore the viability of DNA for future storage needs.
I’ve worked at Micron more than 25 years. During that time I’ve been proud of the innovations that my colleagues and I have spearheaded. Memory and storage won’t sit still. A commitment to innovation and creativity is required to meet demands of the data age. These innovations may not grab a lot of headlines but they are fundamental to enabling continued progress across a range of industries. They will help fuel the next generation of technologies such as self-driving cars, space exploration, artificial intelligence and machine learning, which sounded like science fiction not so long ago.