Energy Modelling of Software for a Hardware Multi-threaded Embedded Microprocessor

Steve Kerrison; Kerstin Eder

doi:10.1145/2700104

Energy Modelling of Software for a Hardware Multi-threaded Embedded Microprocessor

Steve Kerrison, Kerstin Eder

Research output: Contribution to journal › Article (Academic Journal) › peer-review

24 Citations (Scopus)

29 Downloads (Pure)

Abstract

This paper examines a hardware multi-threaded microprocessor and discusses the impact such an architecture has on existing software energy modelling techniques. A framework is constructed for analysing the energy behaviour of the XMOS XS1-L multi-threaded processor and a variation on existing software energy models is proposed, based on analysis of collected energy data. It is shown that by combining execution statistics with sufficient data on the processor’s thread activity and instruction execution costs, a multi-threaded software energy model used with Instruction Set Simulation can yield an average error margin of less than 7%.

Original language	English
Article number	56
Journal	ACM Transactions on Embedded Computing Systems
Volume	14
Issue number	3
DOIs	https://doi.org/10.1145/2700104
Publication status	Published - Apr 2015

Keywords

Computer architecture
Embedded systems
ISA-level energy modeling
Multithreading
Software energy modeling
XMOS XS1 xCORE
energy aware computing

Access to Document

10.1145/2700104

Link to publication in Scopus

Fingerprint Dive into the research topics of 'Energy Modelling of Software for a Hardware Multi-threaded Embedded Microprocessor'. Together they form a unique fingerprint.

2 Finished

ICT-Energy: Co-ordinating Research Efforts of the ICT-Energy Community
Eder, K. I.
1/10/13 → 30/09/16
Project: Research
(ENTRA) Whole-System ENergy TRAnsparency
Eder, K. I.
1/10/12 → 1/01/16
Project: Research

Professor Kerstin I Eder
- Kerstin.Eder@bristol.ac.uk
- Trustworthy Systems Laboratory
- Department of Computer Science - Professor of Computer Science
- Microelectronics
- Bristol Robotics Laboratory
- Systems Centre
Person: Academic , Member

Cite this

@article{c73b692c57a149fda322fa4d35da87d0,

title = "Energy Modelling of Software for a Hardware Multi-threaded Embedded Microprocessor",

abstract = "This paper examines a hardware multi-threaded microprocessor and discusses the impact such an architecture has on existing software energy modelling techniques. A framework is constructed for analysing the energy behaviour of the XMOS XS1-L multi-threaded processor and a variation on existing software energy models is proposed, based on analysis of collected energy data. It is shown that by combining execution statistics with sufficient data on the processor{\textquoteright}s thread activity and instruction execution costs, a multi-threaded software energy model used with Instruction Set Simulation can yield an average error margin of less than 7%.",

keywords = "Computer architecture, Embedded systems, ISA-level energy modeling, Multithreading, Software energy modeling, XMOS XS1 xCORE, energy aware computing",

author = "Steve Kerrison and Kerstin Eder",

year = "2015",

month = apr,

doi = "10.1145/2700104",

language = "English",

volume = "14",

journal = "ACM Transactions on Embedded Computing Systems",

issn = "1539-9087",

publisher = "Association for Computing Machinery (ACM)",

number = "3",

}

TY - JOUR

T1 - Energy Modelling of Software for a Hardware Multi-threaded Embedded Microprocessor

AU - Kerrison, Steve

AU - Eder, Kerstin

PY - 2015/4

Y1 - 2015/4

N2 - This paper examines a hardware multi-threaded microprocessor and discusses the impact such an architecture has on existing software energy modelling techniques. A framework is constructed for analysing the energy behaviour of the XMOS XS1-L multi-threaded processor and a variation on existing software energy models is proposed, based on analysis of collected energy data. It is shown that by combining execution statistics with sufficient data on the processor’s thread activity and instruction execution costs, a multi-threaded software energy model used with Instruction Set Simulation can yield an average error margin of less than 7%.

AB - This paper examines a hardware multi-threaded microprocessor and discusses the impact such an architecture has on existing software energy modelling techniques. A framework is constructed for analysing the energy behaviour of the XMOS XS1-L multi-threaded processor and a variation on existing software energy models is proposed, based on analysis of collected energy data. It is shown that by combining execution statistics with sufficient data on the processor’s thread activity and instruction execution costs, a multi-threaded software energy model used with Instruction Set Simulation can yield an average error margin of less than 7%.

KW - Computer architecture

KW - Embedded systems

KW - ISA-level energy modeling

KW - Multithreading

KW - Software energy modeling

KW - XMOS XS1 xCORE

KW - energy aware computing

UR - http://www.scopus.com/inward/record.url?scp=84929169023&partnerID=8YFLogxK

U2 - 10.1145/2700104

DO - 10.1145/2700104

M3 - Article (Academic Journal)

AN - SCOPUS:84929169023

VL - 14

JO - ACM Transactions on Embedded Computing Systems

JF - ACM Transactions on Embedded Computing Systems

SN - 1539-9087

IS - 3

M1 - 56

ER -

Energy Modelling of Software for a Hardware Multi-threaded Embedded Microprocessor

Abstract

Keywords

Access to Document

ICT-Energy: Co-ordinating Research Efforts of the ICT-Energy Community

(ENTRA) Whole-System ENergy TRAnsparency

Professor Kerstin I Eder

Cite this