Exploring Thermal/Power Optimization for Emerging Computing Platforms

MISSING IMAGE

Material Information

Title:
Exploring Thermal/Power Optimization for Emerging Computing Platforms Think Big, Act Small
Physical Description:
1 online resource (160 p.)
Language:
english
Creator:
Qouneh, Amer M
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Electrical and Computer Engineering
Committee Chair:
LI,TAO
Committee Co-Chair:
FIGUEIREDO,RENATO JANSEN
Committee Members:
LI,XIAOLIN
CHEN,SHIGANG

Subjects

Subjects / Keywords:
datacenter -- efficiency -- heterogeneous -- hpc -- network-on-chip -- optimization -- photonic -- platform -- power -- thermal
Electrical and Computer Engineering -- Dissertations, Academic -- UF
Genre:
Electrical and Computer Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
The explosive growth of large-scale Internet services and the rising popularity of cloud computing have affirmed the role of data centers as the imperative IT infrastructure. To support these services, data centers are equipped with state-of-the-art hardware to maximize performance and improve scalability. However, the power density of these emerging computing platforms has been increasing at a potentially unsustainable rate and causing energy inefficiencies due to thermal effects. In the US alone, data centers consumed 100 billion kWh in the year 2011 at a cost of $7.4 billion. Of that, cooling costs represent 30-50% of the total energy overhead. Improving the energy efficiency of data centers is best tackled at all layers of the computer hierarchy: application, operating system, platform, and micro-architecture layers. However, to achieve maximum efficiency, data centers should also be viewed and modeled at the system level as a single unit. Optimizations at each layer contribute to the overall efficiency of the data center. Furthermore, these optimizations are additive and collectively can achieve significant savings in power and energy consumptions. This work explores thermal and power optimizations at micro-architecture, platform, and data center levels. We characterize the behavior of emerging computing platforms and identify hidden opportunities to maximize power and energy efficiency while meeting performance demands. At the micro-architecture level, we propose techniques to alleviate thermal effects on photonic Networks-on-Chip (NoC) in future multi-core processors. At the platform level, we characterize emerging high performance computing platforms and identify opportunities for saving power and energy. Based on our results, we exploit heterogeneity to improve performance and reduce power budget requirements for servers in cloud data centers. At the data center level, we provide comparative and quantitative analysis of cooling power in both container-based and raised-floor data centers. The goal of our work is to explore power saving techniques across different levels which synergistically mitigate power consumption and improve efficiency of data centers.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Amer M Qouneh.
Thesis:
Thesis (Ph.D.)--University of Florida, 2014.
Local:
Adviser: LI,TAO.
Local:
Co-adviser: FIGUEIREDO,RENATO JANSEN.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2015-05-31

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2014
System ID:
UFE0046698:00001