Energy Optimization for Large-Scale 3D Manycores in the Dark-Silicon Era

In this paper, we study the impact of the idle/dynamic power consumption ratio on the effectiveness of a multi-Vdd/frequency manycore design. We propose a new tool called LVSiM (a Low-Power and Variation-Aware Manycore Simulator) to carry out the experiments. It is a novel manycore simulator targete...

Full description

Bibliographic Details
Main Authors: Sohaib Majzoub, Resve A. Saleh, Imran Ashraf, Mottaqiallah Taouil, Said Hamdioui
Format: Article
Language:English
Published: IEEE 2019-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/8648367/
id doaj-e8838775bc0b4a1d8f668f417739c83e
record_format Article
spelling doaj-e8838775bc0b4a1d8f668f417739c83e2021-03-29T22:50:08ZengIEEEIEEE Access2169-35362019-01-017331153312910.1109/ACCESS.2019.29004778648367Energy Optimization for Large-Scale 3D Manycores in the Dark-Silicon EraSohaib Majzoub0https://orcid.org/0000-0003-3196-2635Resve A. Saleh1Imran Ashraf2Mottaqiallah Taouil3Said Hamdioui4Department of Electrical and Computer Engineering, University of Sharjah, Sharjah, United Arab EmiratesDepartment of Electrical and Computer Engineering, The University of British Columbia, Vancouver, CanadaLaboratory of Computer Engineering, Delft University of Technology, Delft, The NetherlandsLaboratory of Computer Engineering, Delft University of Technology, Delft, The NetherlandsLaboratory of Computer Engineering, Delft University of Technology, Delft, The NetherlandsIn this paper, we study the impact of the idle/dynamic power consumption ratio on the effectiveness of a multi-Vdd/frequency manycore design. We propose a new tool called LVSiM (a Low-Power and Variation-Aware Manycore Simulator) to carry out the experiments. It is a novel manycore simulator targeted towards low-power optimization methods including within-die process and workload variations. LVSiM provides a holistic platform for multi-Vdd/frequency voltage island analysis, optimization, and design. It provides a tool for the early design exploration stage to analyze large-scale manycores with a given number of cores on 3D-stacked layers, network-on-chip communication busses, technology parameters, voltage and frequency values, and power grid parameters, using a variety of different optimization methods. LVSiM has been calibrated with Sniper/McPAT at a nominal frequency, and then the energy-delay-product (EDP) numbers were compared after frequency scaling. The average error is shown to be 10% after frequency scaling, which is sufficient for our purposes. The experiments in this work are carried out for different Idle/Dynamic ratios considering 1260 benchmarks with task sizes ranging from 4000 to 16 000 executing on 3200 cores. The best configurations are shown to produce on average 20.7% to 24.6% EDP savings compared to the nominal configuration. Traditional scheduling methods are used in the nominal configuration with the unused cores switched off. In addition, we show that, as the Idle/Dynamic ratio increases, the multi-Vdd/frequency approach becomes less effective. In the case of a high Idle/Dynamic ratio, the minimum EDP can be achieved through switching off unused cores as opposed to using a multi-Vdd/frequency approach. This conclusion is important, especially in the dark-silicon era, where switching cores on and/or off as needed is a common practice.https://ieeexplore.ieee.org/document/8648367/3D-stacked chipdark-silicondynamic powerenergy-delay-productfrequency scalingidle power
collection DOAJ
language English
format Article
sources DOAJ
author Sohaib Majzoub
Resve A. Saleh
Imran Ashraf
Mottaqiallah Taouil
Said Hamdioui
spellingShingle Sohaib Majzoub
Resve A. Saleh
Imran Ashraf
Mottaqiallah Taouil
Said Hamdioui
Energy Optimization for Large-Scale 3D Manycores in the Dark-Silicon Era
IEEE Access
3D-stacked chip
dark-silicon
dynamic power
energy-delay-product
frequency scaling
idle power
author_facet Sohaib Majzoub
Resve A. Saleh
Imran Ashraf
Mottaqiallah Taouil
Said Hamdioui
author_sort Sohaib Majzoub
title Energy Optimization for Large-Scale 3D Manycores in the Dark-Silicon Era
title_short Energy Optimization for Large-Scale 3D Manycores in the Dark-Silicon Era
title_full Energy Optimization for Large-Scale 3D Manycores in the Dark-Silicon Era
title_fullStr Energy Optimization for Large-Scale 3D Manycores in the Dark-Silicon Era
title_full_unstemmed Energy Optimization for Large-Scale 3D Manycores in the Dark-Silicon Era
title_sort energy optimization for large-scale 3d manycores in the dark-silicon era
publisher IEEE
series IEEE Access
issn 2169-3536
publishDate 2019-01-01
description In this paper, we study the impact of the idle/dynamic power consumption ratio on the effectiveness of a multi-Vdd/frequency manycore design. We propose a new tool called LVSiM (a Low-Power and Variation-Aware Manycore Simulator) to carry out the experiments. It is a novel manycore simulator targeted towards low-power optimization methods including within-die process and workload variations. LVSiM provides a holistic platform for multi-Vdd/frequency voltage island analysis, optimization, and design. It provides a tool for the early design exploration stage to analyze large-scale manycores with a given number of cores on 3D-stacked layers, network-on-chip communication busses, technology parameters, voltage and frequency values, and power grid parameters, using a variety of different optimization methods. LVSiM has been calibrated with Sniper/McPAT at a nominal frequency, and then the energy-delay-product (EDP) numbers were compared after frequency scaling. The average error is shown to be 10% after frequency scaling, which is sufficient for our purposes. The experiments in this work are carried out for different Idle/Dynamic ratios considering 1260 benchmarks with task sizes ranging from 4000 to 16 000 executing on 3200 cores. The best configurations are shown to produce on average 20.7% to 24.6% EDP savings compared to the nominal configuration. Traditional scheduling methods are used in the nominal configuration with the unused cores switched off. In addition, we show that, as the Idle/Dynamic ratio increases, the multi-Vdd/frequency approach becomes less effective. In the case of a high Idle/Dynamic ratio, the minimum EDP can be achieved through switching off unused cores as opposed to using a multi-Vdd/frequency approach. This conclusion is important, especially in the dark-silicon era, where switching cores on and/or off as needed is a common practice.
topic 3D-stacked chip
dark-silicon
dynamic power
energy-delay-product
frequency scaling
idle power
url https://ieeexplore.ieee.org/document/8648367/
work_keys_str_mv AT sohaibmajzoub energyoptimizationforlargescale3dmanycoresinthedarksiliconera
AT resveasaleh energyoptimizationforlargescale3dmanycoresinthedarksiliconera
AT imranashraf energyoptimizationforlargescale3dmanycoresinthedarksiliconera
AT mottaqiallahtaouil energyoptimizationforlargescale3dmanycoresinthedarksiliconera
AT saidhamdioui energyoptimizationforlargescale3dmanycoresinthedarksiliconera
_version_ 1724190732251561984