The Price of Performance

TLDR

CMP was promoted in the late 1990s as an alternative to complex single‑threaded CPUs, and today Google’s infrastructure prioritizes not only performance but also the ability to afford the required computational capacity. The authors’ main goal was to achieve the best commercial workload performance for a given silicon budget. They designed the Piranha system, employing simple RISC‑style cores to maximize thread‑level parallelism. Google’s high‑computational demands have driven a deep understanding of computing cost and a continuous search for hardware/software designs that optimize performance per unit of cost.

Abstract

In the late 1990s, our research group at DEC was one of a growing number of teams advocating the CMP (chip multiprocessor) as an alternative to highly complex single-threaded CPUs. We were designing the Piranha system,1 which was a radical point in the CMP design space in that we used very simple cores (similar to the early RISC designs of the late ’80s) to provide a higher level of thread-level parallelism. Our main goal was to achieve the best commercial workload performance for a given silicon budget. Today, in developing Google’s computing infrastructure, our focus is broader than performance alone. The merits of a particular architecture are measured by answering the following question: Are you able to afford the computational capacity you need? The high-computational demands that are inherent in most of Google’s services have led us to develop a deep understanding of the overall cost of computing, and continually to look for hardware/software designs that optimize performance per unit of cost.