The Effect of Temperature on Amdahl Law in 3D Multicore Era

TL;DR

This paper investigates how temperature constraints impact the performance and scalability of 3D Chip Multiprocessors, revealing thermal limits can hinder scalability even before maximum power is reached, especially with high parallelism.

Contribution

It introduces a thermal perspective to Amdahl's law in 3D CMPs, highlighting the need for alternative architectures to overcome thermal scalability limits.

Findings

3D CMPs reach thermal limits before maximum power

High parallelism causes high peak temperatures

Thermal constraints limit 3D CMP scalability

Abstract

This work studies the influence of temperature on performance and scalability of 3D Chip Multiprocessors (CMP) from Amdahl law perspective. We find that 3D CMP may reach its thermal limit before reaching its maximum power. We show that a high level of parallelism may lead to high peak temperatures even in small scale 3D CMPs, thus limiting 3D CMP scalability and calling for different, in-memory computing architectures.