A New Golden Age for Computer Architecture: History, Challenges, and Opportunities

In the 1980s, Mead and Conway democratized chip design and high-level language programming surpassed assembly language programming, which made instruction set advances viable. Innovations like Reduced Instruction Set Computers (RISC), superscalar, and speculation ushered in a Golden Age of computer architecture, when performance doubled every 18 months. The ending of Dennard Scaling and the recentslowing of Moore’s Law crippled this path; microprocessor performance improved < 10% last year. In addition to poor performance gains, timing attacks on modern microprocessors show they can leak information at high rates.

The ending of Dennard scaling, the slowing of Moore’s law, and the deceleration of performance gains for standard microprocessors are not problems that must be solved but facts that if accepted offer breathtaking opportunities. We believe that high-level, domain-specific languages and architectures, freeing architects from the chains of proprietary instruction sets, and the demand from the public for improved security will usher in a new Golden Age for computer architecture. We envision that domain specific architectures will deliver the same rapid improvement as in the last Golden Age, but this time in cost, energy, and security as well as in performance. One increasingly important target is machine learning, which is dependent in part on novel custom hardware to advance the state of the art of machine learning.

Like the 1980s, the next decade will be exciting for computer architects!