A Measurement-Based Characterization of the Energy Consumption in Data Center Servers

Aroca, Jordi Arjona; Chatzipapas, Angelos; Anta, Antonio Fernández; Mancuso, Vincenzo

doi:10.1109/jsac.2015.2481198

Cited by 29 publications

(12 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The power consumption of the hotspot prototype is analysed for the different measured subsystems, according to the results obtained after post-processing the captured data for the different experiments. While this behaviour was expected for legacy Ethernet devices [46]], it is also observed when NICs are equipped with modern power saving features, such as those defined by the energy efficient Ethernet (EEE; which is the case of the Supermicro servers). This is because energy savings are linked to low and bursty data activity, which supports the use of aggregation techniques (i.e., coalescing) [47].…”

Section: Observed Energy Consumption Resultsmentioning

confidence: 93%

Design, implementation and experimental validation of a 5G energy-aware reconfigurable hotspot

Font-Bach

Bartzoudis

Miozzo

et al. 2018

Computer Communications

View full text Add to dashboard Cite

Flexibility and energy efficiency are considered two principal requirements of future fifth generation (5G) systems. From an architectural point of view, centralized processing and a dense deployment of small cells will play a vital role in enabling the efficient and dynamic operation of 5G networks. In this context, reconfigurable hotspots will provide on-demand services and adapt their operation in accordance to traffic requirements, constituting a vital element of the heterogeneous 5G network infrastructure. In this paper we present a reconfigurable hotspot which is able to flexibly distribute its underlying communication functions across the network, as well as to adapt various parameters affecting the generation of the transmitted signal. The reconfiguration of the hotspot focuses

show abstract

Section: Observed Energy Consumption Resultsmentioning

confidence: 93%

Design, implementation and experimental validation of a 5G energy-aware reconfigurable hotspot

Font-Bach

Bartzoudis

Miozzo

et al. 2018

Computer Communications

View full text Add to dashboard Cite

show abstract

“…Researchers who focus on the performance comparison between the current energy models and the state‐of‐the‐art model of NFV can use CloudSimSDN‐NFV directly to produce relatively accurate results because NFV as the software is still using the same networking resources (switches) and computing resources (physical hosts in cloud data centers) . Power modeling of the virtualized data center can be categorized into analytic power estimation modeling and empirical measurement‐based characterization . In addition, we need to consider other parameters for comprehensive and integrated energy modeling, such as (1) power consumption model for other devices, ie, storage and memory, and (2) temperature flow model between different components and servers.…”

Section: Future Directions and Extensionsmentioning

confidence: 99%

“…25 Power modeling of the virtualized data center can be categorized into analytic power estimation modeling 25 and empirical measurement-based characterization. 31 In addition, we need to consider other parameters for comprehensive and integrated energy modeling, such as (1) power consumption model for other devices, ie, storage and memory, and (2) temperature flow model between different components and servers. Based on the currently available monitoring ability and extension flexibility, many researchers have proposed new solutions for energy modeling.…”

Section: Energy Modelingmentioning

confidence: 99%

CloudSimSDN‐NFV: Modeling and simulation of network function virtualization and service function chaining in edge computing environments

Son

Buyya

2019

Softw Pract Exp

View full text Add to dashboard Cite

Summary Software‐defined networking (SDN) has evolved and brought an innovative paradigm shift in computer networks by utilizing a programmable software controller with open protocols. Network functions, previously served on dedicated hardware, have shifted to network function virtualization (NFV) that enabled functions to be virtualized and provisioned dynamically upon generic hardware. In addition to NFV, edge computing utilizes the edge resources close to end‐users, which can reduce the end‐to‐end service delay and the network traffic volume. Although these innovative technologies gained significant attention from both academia and industry, there are limited tools and simulation frameworks for the effectiveness evaluation in a repeatable and controllable manner. Furthermore, large‐scale experimental infrastructures are expensive to setup and difficult to maintain. Even if they are created, they are not available or accessible for the majority of researchers throughout the world. In this paper, we propose a framework for simulating NFV functionalities in both edge and cloud computing environments. In addition to the basic network functionalities supported by SDN in CloudSimSDN, we added new NFV features, such as virtualized network functions allocation, migration, and autoscaling with the support of corresponding network functionalities, such as flow load balancing, rerouting, and service function chaining (SFC) maintenance. We evaluated our simulation framework with autoscaling and placement policies for SFC in the integrated edge and cloud computing environments. The results demonstrate its effectiveness in measuring and evaluating the end‐to‐end delay, response time, resource utilization, network traffic, and power consumption with different algorithms in each scenario.

show abstract

“…Although the processor accounts for most of the power consumption of a server, there are some other components such as hard disks and network component, which also consume power [51]. To estimate the cooling demand more accurately, it is assumed that the power consumption of hard disks and network components are 80 W and 20 W respectively, and each server contains 20 processors.…”

Section: Scenariomentioning

confidence: 99%

Investigation on the Energy Saving Potential of Using a Novel Dew Point Cooling System in Data Centres

Wang

et al. 2017

Energies

View full text Add to dashboard Cite

Information technology (IT) has brought significant changes in people’s lives. As an important part of the IT industry, data centres (DCs) have been rapidly growing in both the number and size over the past 40 years. Around 30% to 40% of electricity consumption in DCs is used for space cooling, thus leading to very inefficient DC operation. To identify ways to reduce the energy consumption for space cooling and increase the energy efficiency of DCs’ operation, a dedicated investigation into the energy usage in DCs has been undertaken and a novel high performance dew point cooling system was introduced into a DC operational scheme. Based on the cooling load in DCs, a case study was carried out to evaluate the energy consumptions and energy usage effectiveness when using the novel dew point cooling system in different scales of DCs in various climates. It was found that by using the novel dew point cooling system, for 10 typical climates a DC can have a much lower power usage effectiveness (PUE) of 1.10 to 1.22 compared to that of 1.7 to 3.7 by using existing traditional cooling systems, leading to significantly increased energy efficiency of the DC operation. In addition, the energy performance by managing the cooling air supply at the different levels in DCs, i.e., room, row and rack level, was simulated by using a dynamic computer model. It was found that cooling air supply at rack level can provide a higher energy efficiency in DCs. Based on the above work, the energy saving potential in DCs was conducted by comparing DCs using an the novel dew point cooling system and the optimum management scheme for the cooling air supply to that using traditional air cooling systems and the same supply air management. Annual electricity consumptions for the two cases were given. It was found that by using the novel dew point cooling system and optimum management system for the cooling air supply, an 87.7~91.6% electricity consumption saving for space cooling in DCs could be achieved in 10 typical cities at 10 selected climatic conditions.

show abstract

A Measurement-Based Characterization of the Energy Consumption in Data Center Servers

Cited by 29 publications

References 22 publications

Design, implementation and experimental validation of a 5G energy-aware reconfigurable hotspot

Design, implementation and experimental validation of a 5G energy-aware reconfigurable hotspot

CloudSimSDN‐NFV: Modeling and simulation of network function virtualization and service function chaining in edge computing environments

Investigation on the Energy Saving Potential of Using a Novel Dew Point Cooling System in Data Centres

Contact Info

Product

Resources

About