Summary of John Hennessy and David Patterson 2017 ACM A.M. Turing Award Lecture

Summary of Key Points from the 2017 ACM A.M. Turing Award Lecture by John Hennessy and David Patterson

Introduction

Vicki Hanson, president of ACM, introduces the Turing Award and its significance in recognizing major contributions to computing.
John L. Hennessy and David A. Patterson are honored for their foundational work on energy-efficient RISC-based processors, which have significantly impacted mobile computing and the Internet of Things (IoT).

Historical Context of Computer Architecture

The lecture begins with a historical overview of computer architecture, focusing on the evolution from complex instruction set computing (CISC) to reduced instruction set computing (RISC).
IBM's early challenges in standardizing instruction sets led to innovations like microprogramming.
The introduction of RISC architecture by Patterson and Hennessy in the 1980s revolutionized microprocessor design, emphasizing simpler, more efficient instruction sets.
The success of RISC processors is evident, as they constitute 99% of the 16 billion microprocessors produced annually.

Current Challenges in Computer Architecture

Hennessy discusses the end of Dennard scaling and Moore's Law, which have shifted design constraints towards power efficiency and energy consumption.
The performance improvements in processors have slowed down significantly, requiring a rethink of architectural approaches.
The security landscape in computing is also highlighted as a critical area needing attention, with historical lessons about security protocols and the need for better collaboration between hardware and software communities.

Future Directions

The speakers advocate for a shift towards domain-specific architectures, which can optimize performance for specific applications rather than trying to create a one-size-fits-all solution.
They emphasize the importance of co-designing hardware and software to achieve better performance and efficiency.
The RISC-V initiative is presented as a promising open architecture that allows for community-driven development and innovation, contrasting with proprietary architectures.

Agile Hardware Development

The lecture introduces the concept of agile hardware development, inspired by agile software methodologies, allowing for rapid prototyping and iteration in chip design.
The use of Cloud-based FPGAs (Field-Programmable Gate Arrays) is highlighted as a way to democratize hardware development, making it more accessible to researchers and smaller teams.

Conclusion

Hennessy and Patterson conclude that we are entering a new Golden Age of computer architecture, where opportunities for innovation abound due to the challenges posed by the end of traditional scaling methods. They encourage collaboration across disciplines to harness the full potential of emerging technologies and methodologies.

Methodology and Key Takeaways

Historical Analysis: Review the evolution of computer architecture from CISC to RISC.
Addressing Current Challenges:
- Acknowledge the end of Dennard scaling and Moore's Law.
- Focus on energy efficiency and security.
Embrace Domain-Specific Architectures: Shift towards architectures tailored for specific applications.
Agile Development: Adopt agile methodologies in hardware design to enhance innovation and responsiveness.
Open Architectures: Support initiatives like RISC-V to foster community engagement and collaboration.

Speakers

John L. Hennessy: Former president of Stanford University, co-founder of MIPS Computer Systems, and current Chairman of the Board of Alphabet Inc.
David A. Patterson: Retired professor at the University of California, Berkeley, and distinguished engineer at Google.
Vicki Hanson: President of ACM, who introduced the lecture.

Notable Quotes

— 02:45 — « The work led by Hennessy at Stanford and Patterson at Berkeley would result in a systematic, quantitative approach to designing faster, lower power and reduced instruction set microprocessors. »

— 03:38 — « The second computer architect to get a Turing Award, Fred Brooks, played a big role in that effort. »

— 36:42 — « We’ve got to find new ways to think about how to use the capability that we have more efficiently. »

— 37:30 — « We’re entering a new Golden Age. »

— 37:50 — « This is a great time to be an architect. »