An Energy-Aware Real-Time VM-Provisioning Framework For Heterogeneous Data Centres

2021 ◽  
Author(s):  
Salam Ismaeel
Keyword(s):  
Energy Consumption ◽  
Real Time ◽  
Power Efficiency ◽  
Large Scale ◽  
Virtual Machines ◽  
Heterogeneous Data ◽  
Energy Aware ◽  
Time Dynamic ◽  
Data Centre ◽  
Data Centres

<div>Increasing power efficiency is one of the most important operational factors for any data centre providers. In this context, one of the most useful approaches is to reduce the number of utilized Physical Machines (PMs) through optimal distribution and re-allocation of Virtual Machines (VMs) without affecting the Quality of Service (QoS). Dynamic VMs provisioning makes use of monitoring tools, historical data, prediction techniques, as well as placement algorithms to improve VMs allocation and migration. Consequently, the efficiency of the data centre energy consumption increases.</div><div>In this thesis, we propose an efficient real-time dynamic provisioning framework to reduce energy in heterogeneous data centres. This framework consists of an efficient workload preprocessing, systematic VMs clustering, a multivariate prediction, and an optimal Virtual Machine Placement (VMP) algorithm. Additionally, it takes into consideration VM and user behaviours along with the existing state of PMs. The proposed framework consists of a pipeline successive subsystems. These subsystems could be used separately or combined to improve accuracy, efficiency, and speed of workload clustering, prediction and provisioning purposes.<br></div><div>The pre-processing and clustering subsystems uses current state and historical workload data to create efficient VMs clusters. Efficient VMs clustering include less consumption resources, faster computing and improved accuracy. A modified multivariate Extreme Learning Machine (ELM)-based predictor is used to forecast the number of VMs in each cluster for the subsequent period. The prediction subsystem takes users’ behaviour into consideration to exclude unpredictable VMs requests.<br></div><div>The placement subsystem is a multi-objective placement algorithm based on a novel Machine Condition Index (MCI). MCI represents a group of weighted components that is inclusive of data centre network, PMs, storage, power system and facilities used in any data centre. In this study it will be used to measure the extent to which PM is deemed suitable for handling the new and/or consolidated VM in large scale heterogeneous data centres. It is an efficient tool for comparing server energy consumption used to augment the efficiency and manageability of data centre resources.</div><div> The proposed framework components separately are tested and evaluated with both synthetic and realistic data traces. Simulation results show that proposed subsystems can achieve efficient results as compared to existing algorithms. <br></div>

An Energy-Aware Real-Time VM-Provisioning Framework For Heterogeneous Data Centres

2021 ◽  
Author(s):  
Salam Ismaeel
Keyword(s):  
Energy Consumption ◽  
Real Time ◽  
Power Efficiency ◽  
Large Scale ◽  
Virtual Machines ◽  
Heterogeneous Data ◽  
Energy Aware ◽  
Time Dynamic ◽  
Data Centre ◽  
Data Centres

<div>Increasing power efficiency is one of the most important operational factors for any data centre providers. In this context, one of the most useful approaches is to reduce the number of utilized Physical Machines (PMs) through optimal distribution and re-allocation of Virtual Machines (VMs) without affecting the Quality of Service (QoS). Dynamic VMs provisioning makes use of monitoring tools, historical data, prediction techniques, as well as placement algorithms to improve VMs allocation and migration. Consequently, the efficiency of the data centre energy consumption increases.</div><div>In this thesis, we propose an efficient real-time dynamic provisioning framework to reduce energy in heterogeneous data centres. This framework consists of an efficient workload preprocessing, systematic VMs clustering, a multivariate prediction, and an optimal Virtual Machine Placement (VMP) algorithm. Additionally, it takes into consideration VM and user behaviours along with the existing state of PMs. The proposed framework consists of a pipeline successive subsystems. These subsystems could be used separately or combined to improve accuracy, efficiency, and speed of workload clustering, prediction and provisioning purposes.<br></div><div>The pre-processing and clustering subsystems uses current state and historical workload data to create efficient VMs clusters. Efficient VMs clustering include less consumption resources, faster computing and improved accuracy. A modified multivariate Extreme Learning Machine (ELM)-based predictor is used to forecast the number of VMs in each cluster for the subsequent period. The prediction subsystem takes users’ behaviour into consideration to exclude unpredictable VMs requests.<br></div><div>The placement subsystem is a multi-objective placement algorithm based on a novel Machine Condition Index (MCI). MCI represents a group of weighted components that is inclusive of data centre network, PMs, storage, power system and facilities used in any data centre. In this study it will be used to measure the extent to which PM is deemed suitable for handling the new and/or consolidated VM in large scale heterogeneous data centres. It is an efficient tool for comparing server energy consumption used to augment the efficiency and manageability of data centre resources.</div><div> The proposed framework components separately are tested and evaluated with both synthetic and realistic data traces. Simulation results show that proposed subsystems can achieve efficient results as compared to existing algorithms. <br></div>

scholarly journals MeReg: Managing Energy-SLA Tradeoff for Green Mobile Cloud Computing

2017 ◽  
Vol 2017 ◽  
pp. 1-11 ◽  
Author(s):  
Rahul Yadav ◽  
Weizhe Zhang
Keyword(s):  
Cloud Computing ◽  
Energy Consumption ◽  
Large Scale ◽  
Data Centers ◽  
Mobile Cloud Computing ◽  
Virtual Machines ◽  
Mobile Cloud ◽  
Energy Aware ◽  
Cpu Utilization ◽  
Selection Algorithms

Mobile cloud computing (MCC) provides various cloud computing services to mobile users. The rapid growth of MCC users requires large-scale MCC data centers to provide them with data processing and storage services. The growth of these data centers directly impacts electrical energy consumption, which affects businesses as well as the environment through carbon dioxide (CO2) emissions. Moreover, large amount of energy is wasted to maintain the servers running during low workload. To reduce the energy consumption of mobile cloud data centers, energy-aware host overload detection algorithm and virtual machines (VMs) selection algorithms for VM consolidation are required during detected host underload and overload. After allocating resources to all VMs, underloaded hosts are required to assume energy-saving mode in order to minimize power consumption. To address this issue, we proposed an adaptive heuristics energy-aware algorithm, which creates an upper CPU utilization threshold using recent CPU utilization history to detect overloaded hosts and dynamic VM selection algorithms to consolidate the VMs from overloaded or underloaded host. The goal is to minimize total energy consumption and maximize Quality of Service, including the reduction of service level agreement (SLA) violations. CloudSim simulator is used to validate the algorithm and simulations are conducted on real workload traces in 10 different days, as provided by PlanetLab.

scholarly journals Energy management in sensor networks

2012 ◽  
Vol 370 (1958) ◽  
pp. 52-67 ◽  
Author(s):  
John A. Stankovic ◽  
Tian He
Keyword(s):  
Energy Consumption ◽  
Sensor Networks ◽  
Energy Management ◽  
Large Scale ◽  
Energy Savings ◽  
Time Synchronization ◽  
Media Access Control ◽  
Media Access ◽  
Energy Aware ◽  
Holistic View

This paper presents a holistic view of energy management in sensor networks. We first discuss hardware designs that support the life cycle of energy, namely: (i) energy harvesting, (ii) energy storage and (iii) energy consumption and control. Then, we discuss individual software designs that manage energy consumption in sensor networks. These energy-aware designs include media access control, routing, localization and time-synchronization. At the end of this paper, we present a case study of the VigilNet system to explain how to integrate various types of energy management techniques to achieve collaborative energy savings in a large-scale deployed military surveillance system.

Secure Cloud Auditability for Virtual Machines by Adaptive Characterization Using Machine Learning Methods

2021 ◽  
Vol 4 (3) ◽  
pp. 59-75
Author(s):  
Shesagiri Taminana ◽  
◽  
Lalitha Bhaskari ◽  
Arwa Mashat ◽  
Dragan Pamučar ◽  
...  
Keyword(s):  
Time Complexity ◽  
Virtual Machines ◽  
Application Development ◽  
Predictive Regression ◽  
Data Centre ◽  
Machine Learning Methods ◽  
Highly Sensitive ◽  
Increasing Demand ◽  
Application Developers ◽  
Data Centres

With the Present days increasing demand for the higher performance with the application developers have started considering cloud computing and cloud-based data centres as one of the prime options for hosting the application. Number of parallel research outcomes have for making a data centre secure, the data centre infrastructure must go through the auditing process. During the auditing process, auditors can access VMs, applications and data deployed on the virtual machines. The downside of the data in the VMs can be highly sensitive and during the process of audits, it is highly complex to permits based on the requests and can increase the total time taken to complete the tasks. Henceforth, the demand for the selective and adaptive auditing is the need of the current research. However, these outcomes are criticised for higher time complexity and less accuracy. Thus, this work proposes a predictive method for analysing the characteristics of the VM applications and the characteristics from the auditors and finally granting the access to the virtual machine by building a predictive regression model. The proposed algorithm demonstrates 50% of less time complexity to the other parallel research for making the cloud-based application development industry a safer and faster place.

Vertical Integration Between Providers With Possible Cloud Migration

2018 ◽  
pp. 1164-1173 ◽  
Author(s):  
Aleksandra Kostic-Ljubisavljevic ◽  
Branka Mikavica
Keyword(s):  
Large Scale ◽  
Virtual Machines ◽  
Wholesale Price ◽  
Rejection Rate ◽  
Large Scale Data ◽  
Cloud Migration ◽  
Computational Resources ◽  
Data Centres ◽  
Charging Strategy

All vertically integrated participants in content provisioning process are influenced by bandwidth requirements. Provisioning of self-owned resources that satisfy peak bandwidth demand leads to network underutilization and it is cost ineffective. Under-provisioning leads to rejection of customers' requests. Vertically integrated providers need to consider cloud migration in order to minimize costs and improve Quality of Service and Quality of Experience of their customers. Cloud providers maintain large-scale data centres to offer storage and computational resources in the form of Virtual Machines instances. They offer different pricing plans: reservation, on-demand and spot pricing. For obtaining optimal integration charging strategy, Revenue Sharing, Cost Sharing, Wholesale Price is applied frequently. The vertically integrated content provider's incentives for cloud migration can induce significant complexity in integration contracts, and consequently improvements in costs and requests' rejection rate.

RSVP-TE Bilateral-Recursive Region Re-Routing Crankback Mechanism for Large-Scale Optical Networks

2019 ◽  
Vol 0 (0) ◽  
Author(s):  
Panke Qin ◽  
Jiawei Wang ◽  
Jingru Wu
Keyword(s):  
Real Time ◽  
Optical Networks ◽  
Traffic Engineering ◽  
Large Scale ◽  
Setup Time ◽  
Multiprotocol Label Switching ◽  
Transmission Scheme ◽  
Internet Applications ◽  
Time Dynamic ◽  
Reservation Protocol

AbstractCloud computing services and real-time Internet applications have spawned a large variety of potential requirements for quality of service (QoS), especially the latency and connection setup time. However, with the optical networks develop toward to larger scale, wider coverage and more users access, conventional resource reservation protocol traffic engineering (RSVP-TE) signal hop by hop transmission scheme cannot meet the requirements of these new applications for real-time dynamic services and fast restoration with long propagation delays. This paper proposes a novel RSVP-TE bilateral-recursive region re-routing crankback mechanism (BRCB) base on distributed path computation element (PCE) for generalized multiprotocol label switching (GMPLS) optical networks. In this mechanism, the backtracking nodes re-route and update the region routing paths which bypass the crankback and re-routing failure nodes when crankback occurs. It not only reduces the influencing factors of the scale of network, signaling crankback position and frequency to path connection setup time, but also avoids the backtracking of teardown messages.

Addressing Thermal Challenges in Design of Data Centres

2010 ◽  
Author(s):  
Babak Fakhim ◽  
Srinarayana Nagarathinam ◽  
Simon Wong ◽  
Masud Behnia ◽  
Steve Armfield
Keyword(s):  
Energy Consumption ◽  
Hot Spots ◽  
Heat Dissipation ◽  
Field Measurements ◽  
Temperature Fields ◽  
Power Requirement ◽  
Data Centre ◽  
It Systems ◽  
Improved Design ◽  
Data Centres

Aggregation of small networking hardware has led to an ever increasing power density in data centres. The energy consumption of IT systems is continuing to rise substantially owing to the demands of electronic information and storage requirements. Energy consumption of data centres can be severely and unnecessarily high due to inadequate localised cooling and densely packed rack layouts. However, as heat dissipation in data centres rises by orders of magnitude, inefficiencies such as air recirculation causing hot spots, leading to flow short-circuiting will have a significant impact on the thermal manageability and energy efficiency of the cooling infrastructure. Therefore, the thermal management of high-powered electronic components is a significant challenge for cooling of data centres. In this project, an operational data centre has been studied. Field measurements of temperature have been performed. Numerical analysis of flow and temperature fields is conducted in order to evaluate the thermal behaviour of the data centre. A number of undesirable hot spots have been identified. To rectify the problem, a few practical design solutions to improve the cooling effectiveness have been proposed and examined to ensure a reduced air-conditioning power requirement. Therefore, a better understanding of the cooling issues and the respective proposed solutions can lead to an improved design for future data centres.

DISSOLVED CLOUDS: ERICSSON'S VAUDREUIL DATA CENTRE AND INFRASTRUCTURAL ABANDONMENT

2020 ◽  
Author(s):  
Julia Velkova ◽  
Patrick Brodie
Keyword(s):  
Large Scale ◽  
Qualitative Interviews ◽  
Cloud Infrastructure ◽  
Corporate Strategies ◽  
Data Centre ◽  
Large Scale Data ◽  
Accelerated Growth ◽  
Data Centres ◽  
Scale Data ◽  
Do So

The past decade has seen the accelerated growth and expansion of large-scale data centre operations across the world to support emerging consumer and business data and computation needs. These buildings, as infrastructures responsive to changing global economic and technological terrain, are increasingly modular, and must be built out rapidly. However, these conditions also mean that their paths to obsolescence are shortened, their lifespans dependent on shifting corporate strategies and advances in consumer technology. This paper theorises and empirically explores material, infrastructural abandonment that emerges in this process of data centre construction across different geographical contexts. To do so, we analyse the socio-material construction of an international network of large-scale data centres by global telecom giant Ericsson, and the abrupt abandonment and suspension of one of its nodes in Vaudreuil, Québec in 2017 after only nine months of operation. Employing autoethnography, site visits, and qualitative interviews with data centre architects and staff in Sweden and Canada, we argue that the ruins of abandoned 'cloud' infrastructure represent the disjunction between the 'promise' of digital infrastructure for local communities and the market interests of digital companies. With its focus, the paper takes ruination and discard as perspectives through which to understand the complexity of emergent datafied futures and the socio-technical reshaping of internet infrastructures.

Service Level Agreement Conformance in Virtual Machine Migration with an Improved Energy and Performance Management Method

2020 ◽  
Vol 17 (9) ◽  
pp. 3904-3906
Author(s):  
Susmita J. A. Nair ◽  
T. R. Gopalakrishnan Nair
Keyword(s):  
Energy Consumption ◽  
Performance Management ◽  
Large Scale ◽  
Data Centers ◽  
Virtual Machines ◽  
Service Level Agreement ◽  
Service Level ◽  
Performance Degradation ◽  
Virtual Machine Migration ◽  
And Performance

Increasing demand of computing resources and the popularity of cloud computing have led the organizations to establish of large-scale data centers. To handle varying workloads, allocating resources to Virtual Machines, placing the VMs in the most suitable physical machine at data centers without violating the Service Level Agreement remains a big challenge for the cloud providers. The energy consumption and performance degradation are the prime focus for the data centers in providing services by strictly following the SLA. In this paper we are suggesting a model for minimizing the energy consumption and performance degradation without violating SLA. The experiments conducted have shown a reduction in SLA violation by nearly 10%.

Sign in / Sign up

Export Citation Format

Share Document