Back to search

Democritus University of Thrace (DUTH)

Design and hardware implementation of unconventional computing systems for networks analysis and data center management

Abstract

dc:description

The high power consumption of current Data Centers is one of the most challenging issues in the contemporary IT ecosystem. According to Greenpeace’s Make IT Green report, it is estimated that the global demand for electricity from Data Centers was around 330bns kWh in 2007 (almost the same amount of electricity consumed by UK). This demand in power consumption is projected to more than triple by 2020 (more than 1000bns kWh). The rise of cloud computing and other emerging web applications has increased significantly the network traffic inside the Data Centers and led to a significant increase of the power consumption. The increase on the power consumption results also to increased carbon emissions. In 2002, the global Data Center footprint, including equipment use and embodied carbon, was 76 MtCO2e (Million metric tons of carbon dioxide equivalent) emissions and this is expected to more than triple by 2020 and reach up to 259 MtCO2e – making it the fastest-growing contributor to the ICT sector’s carbon footprint, at 7% in relative terms. Reducing significantly the power consumption is impossible without a dramatic improvement in energy efficiency, decreasing the computation and communication energy consumption per bit. Therefore, more energy efficient processors, storage elements, switches and memories are required that will provide increased processing power with reduced energy consumption. Currently, almost all of the web servers are based on high performance x86 architectures which are not optimized in terms of power. The increasingly energy consumption of Data Centers provided motivation for this PhD dissertation. An additional factor was the low utilization of alternative computing methods that require interdisciplinary scientific field backgrounds. An unconventional medium used as inspiration for this PhD thesis is the plasmodium of Physarum polycephalum, a multi-nuclei cell, a substrate that is alive, seeks for nutrients and struggles to survive in a harsh environment. A model was produced after analyzing the plasmodium of P. polycephalum that designed networks which were tested against already existing networks. Moreover, based on the results and theoretically defined graphs, the model was optimized and provided more efficient networks. Thus, the same model was utilized to design a data tree for a Wireless Sensor Network (WSN) that is employed to monitor the temperature in a Data Center and control its supporting systems. On the other hand, rather conventional ways to provide more efficient Data Centers were studied. Specifically, the allocation of hardware resources to distributed computation entities using Cellular Automata and Game Theory was investigated. Furthermore, the implementation of computations in hardware was examined in order to accelerate the computations on a Data Center and therefore, economize energy.

Degree

thesis:*
Grantor dc:publisher
Democritus University of Thrace (DUTH)
Year dc:date
2016

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Tsompanas, Michail-Antisthenis
  • Τσομπανάς, Μιχαήλ-Αντισθένης

Subjects

dc:subject × 14

Rights

Language dc:language
eng

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/39877

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Tsompanas, Michail-Antisthenis; Τσομπανάς, Μιχαήλ-Αντισθένης. Design and hardware implementation of unconventional computing systems for networks analysis and data center management. Democritus University of Thrace (DUTH), 2016. http://hdl.handle.net/10442/hedi/39877