A new golden age for computer architecture

By John L. Hennessy, David A. Patterson. Communications of the ACM, 2019

691 citations recorded by the source corpus.

Innovations like domain-specific hardware, enhanced security, open instruction sets, and agile chip development will lead the way.

Subfields: Electrical and Electronic Engineering, Hardware and Architecture

View Paper: https://www.semanticscholar.org/paper/a3967f1a6f2bc027c18538f3c3fc3cdc09c5d298

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Electrical and Electronic Engineering

In Hardware and Architecture

Citations in the corpus

Benchmarking TPU, GPU, and CPU Platforms for Deep Learning (2019, background): “As the end of Dennard scaling and Moore’s law has slowed the performance improvement of general-purpose microprocessors [23], the design of domain-speciﬁc hardware is becoming more and more relevant.”
Learning Execution through Neural Code Fusion (2019, methodology): “The end of this exponential scaling has enormous ramifications for computing [18] since the demand for compute has simultaneously grown exponentially, relying on Moore’s Law to compensate [34].”
What serverless computing is and should become (2021, background): “14 While there are concerns today about serverless security, we believe that a careful design could in fact make it easier for application developers to secure their software against external attackers.”
Cloud Programming Simplified: A Berkeley View on Serverless Computing (2019, background): “Fallacy Since serverless computing programming is in high-level languages like Python, it’s easy to port applications between serverless computing providers.”
Ten Lessons From Three Generations Shaped Google’s TPUv4i : Industrial Product (2021, background): “It shows the evolution of an architecture family, as production experience has informed new designs beyond the familiar issues of sluggish CPUs and a diminishing Moore’s Law [14].”
Survey of Machine Learning Accelerators (2020, background): “This includes an explosion of innovation in ML processors and accelerators [4], [5].”
AccPar: Tensor Partitioning for Heterogeneous Deep Learning Accelerators (2020, background): “Domain-specific computing architectures [156–159] are considered as a promising solution to accommodate the evergrowing intensive computing in various deep learning applications.”
There’s plenty of room at the Top: What will drive computer performance after Moore’s law? (2020, background): “Hennessy and Patterson ( 45 ) foresee a move from general-purpose toward domain-specific architectures that run small compute-intensive kernels of larger systems for tasks such as object recognition or speech understanding.”