Design and Evaluation of Dynamically Specialized Datapaths with the DySER Architecture

Abstract

Due to limits in technology scaling, energy efficiency of logic devices is decreasing in successive generations. To provide continued performance improvements without increasing power, regardless of the sequential or parallel nature of the application, microarchitectural energy efficiency must improve. We propose Dynamically Specialized Execution to improve the energy efficiency of general purpose programmable processors. The key insights of this work are the following. First, applications execute in phases and these phases can be determined by creating a path-tree of basic-blocks rooted at the inner-most loop. Second, specialized datapaths corresponding to these path-trees can be constructed by interconnecting a set of heterogeneous computation units with a circuit-switched network, which we refer to as DySER blocks. These blocks can be easily integrated with a conventional processor.

A synthesized RTL implementation using an industry 55nm technology library shows a 64-functional-unit DySER block occupies approximately the same area as a 64 KB single-ported SRAM and can execute at 2 GHz. Using the GCC compiler, we identify path-trees and evaluate the PARSEC, SPEC and Parboil benchmarks suites, using our extensions for mapping code to DySER. Our results show that in most cases two DySER blocks can achieve the same performance (within 5%) as having a specialized block for each path-tree. A 64-FU DySER block can cover 12% to 100% of the dynamically executed instruction stream. When integrated with a dual-issue out-of-order processor, two DySER blocks provide geometric mean speedup of 2.1X (1.15X to 10X), and geometric mean energy reduction of 40% (up to 70%), and 60% energy reduction if no performance improvement is required.