Seven nanometers. That’s slightly longer than the 2.5 nm circumference of a DNA double helix. And 100,000 times smaller than the width of a human hair. It’s also the miniscule dimension that future semiconductors must push past to keep up with today’s computing demands being put on cloud, big data and cognitive systems.
IBM’s five-year $3 billion investment in research and early-stage development aims to take semiconductors beyond seven nanometers – and beyond silicon, today’s ubiquitous chip material that’s fast becoming the bottleneck to next generation computers required for a new era of computing.
Beyond 7 nanometers
To keep cloud computing’s promise of better, more-efficient technology delivery – whether for enterprise-level banking, or just downloading apps on your mobile device – bandwidth, memory, and the communication within chips and systems between the machines that make the cloud work needs to scale. Today’s 22 nm, and future 14 nm and 10 nm technologies, will run out of steam before long.
Getting to seven nanometers (and beyond) by the end of the decade will require not just new chip designs, but new tools, such as nanowires and nanochisels to manufacture them.
“The question is not if we will introduce seven-nanometer technology into manufacturing, but rather how, when, and at what cost?” said John Kelly, senior vice president, IBM Research. “IBM engineers and scientists, along with our partners, are well suited for this challenge and are already working on the materials science and device engineering required to meet the demands of the emerging system requirements for cloud, big data, and cognitive systems. This new investment will ensure that we produce the necessary innovations to meet these challenges.”
Beyond silicon
This investment is also focused on developing alternative technologies for when silicon can’t meet the nanoscale needs of future systems. And seven nanometers appears to be silicon’s limit. Our scientists are exploring everything from carbon nanotubes, to silicon nanophotonics and graphene as that next chip material.
IBM Research is already making breakthroughs in these areas, such as demonstrating the world’s first graphene based integrated circuit receiver front end for wireless communications; earning a Nobel Prize for the scanning tunneling microscope; and developing a non-von Neumann architecture neurosynaptic chip. This “neuron-like” technology emulates the brain’s computing efficiency, size and power usage. Long term, a neurosynaptic system will have billions of neurons using its own programming language – all while consuming just a kilowatt of power and occupying less than two liters of volume.
IBM is also working in superconducting qubit-based quantum computing. And while still a “fundamental science” today, when this technology matures, it will solve problems that are today either impossible or impractical to apply conventional machines to. The team recently demonstrated the first experimental realization of parity check with three superconducting qubits, an essential building block for one type of quantum computer.
This investment in new materials and techniques promise to not only keep the status quo of Moore’s Law for years to come, but to also deliver these new neuromorphic and quantum computational approaches. It reflects IBM’s commitment to leading advanced semiconductor research for future high-performance systems and cognitive computing.
