Thomas Sterling has retracted his prediction that we will never reach ZettaFLOP computers. He now predicts zettaFLOPS can be achieved in less than 10 years if innovations in non-von Neumann architecture can be scaled. With a change to cryogenic technologies, we can reach yottaFLOPS by 2030.
The world is currently on the verge of ExaFLOP supercomputers. The USA just revealed a 200 petaFLOP supercomputer. China will soon complete three supercomputers that will be close to ExaFLOP and one or more could reach ExaFLOP performance.
Name Unit Value kiloFLOPS kFLOPS 10^3 megaFLOPS MFLOPS 10^6 gigaFLOPS GFLOPS 10^9 teraFLOPS TFLOPS 10^12 petaFLOPS PFLOPS 10^15 exaFLOPS EFLOPS 10^18 zettaFLOPS ZFLOPS 10^21 yottaFLOPS YFLOPS 10^24 brontoFLOPS BFLOPS 10^27 GEOFLOPS GeFLOPS 10^30
Many see quantum computing, neuromorphic computing, and optical computing as technologies which could disrupt high-performance computing.
Niobium Josephson Junction-based technologies cooled to four Kelvins can operate beyond 100 and 200 GHz and has slowly evolved over two or more decades. Quantum annealing is performed at 40 milli-Kelvins or lower. The four Kelvin requirement is easy in comparison. However, latencies measured in cycles grow proportionally with clock rate and superconducting supercomputing must take a very distinct form from typical von Neumann cores. Sterling is not putting forward a controversial position.
Possible alternative non-von Neumann architectures could address the latency challenge. Cellular automata and data flow could be used, but problems will need to be overcome. The post-exascale era has a lot of possibilities.
There was a Cordis research project on non-von Neumann architectures. Reconfigurable non-von-Neumann Accelerators.
Nextbigfuture believes eventual nanocomputing will deliver energy efficient computers beyond YottaFLOPS
At the same ten megawatt power level as some supercomputers, molecular nanocomputer supercomputers could achieve YottaFlop, BrontoFlops or even GeoFlop compute levels.
Ralph Merkle, Robert Freitas and others have a theoretical design for a molecular mechanical computer that would be 100 billion times more energy efficient than the most energy efficient conventional green supercomputer. Removing the need for gears, clutches, switches, springs makes the design easier to build.
Existing designs for mechanical computing can be vastly improved upon in terms of the number of parts required to implement a complete computational system. Only two types of parts are required: Links, and rotary joints. Links are simply stiff, beam-like structures. Rotary joints are joints that allow rotational movement in a single plane
A molecular model of a diamond-based lock, ¾ view
So a technological brute force acceleration looks likely in the 15 to 35 year timeframe. We will at least have some improvements on deep learning and reinforcement learning. Substantial trillion+ qubit general purpose quantum computers and all optical computers will also likely be available.