Energy and Thermal Aware Scheduling in Data Centers

Zomaya, Albert Y.; Lee, Young Choon

doi:10.1002/9781118342015.ch11

Cited by 3 publications

(2 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The point of optimal utilisation hence becomes the factor that determines if new requests could be accommodated to the same physical machine. Some similar comments on how poor utilisation impacts energy were also made in a publication by Dhiman et al (2012). They make some interesting discussion on effects of consolidation of Virtual Machines (VMs) across physical machines.…”

Section: Challenges For Network Based Energy Optimisationsmentioning

confidence: 81%

The Need for Traffic Based Virtualisation Management for Sustainable Clouds

Voderhobli

2015

International Journal of Organizational and Collective Intelligence

View full text Add to dashboard Cite

The subscription based model of cloud computing has allowed for users to migrate their processes to off-site network facilities. One of the motivations for deployment of cloud based services is to promote sustainability by reducing green-house emissions at a local level. Although some might argue that this model saves a lot of power at local network facilities, the problem of energy crisis caused by ICT is never-ending. Today, data-centres are bee-hives of exascale computing and high network dependant processing. Work-load on the cloud directly contributes to energy consumption to an extent that currently IT clouds are some of the worst contributors to CO2 emissions. This paper discusses why traffic-management on the cloud is vital to make it more power efficient and how it can be achieved by gathering live network statistics. The discussion fits with the context of “emerging clouds” as thought needs to be given on how to apply energy efficient schemes at various points including at the communication level.

show abstract

Section: Challenges For Network Based Energy Optimisationsmentioning

confidence: 81%

The Need for Traffic Based Virtualisation Management for Sustainable Clouds

Voderhobli

2015

International Journal of Organizational and Collective Intelligence

View full text Add to dashboard Cite

show abstract

“…These improvements have allowed the direct application of this work to networked data centers and green-energy-aware routing. 20 For embedded systems, SEELab research has investigated context-aware energy management of mobile devices. Using fuzzy inference, Android applications can adjust operating parameters based on the surrounding environment to reduce active power consumption.…”

Section: Increasing Green Energy Efficiencymentioning

confidence: 99%

"Green Machine" Intelligence: Greening and Sustaining Smart Grids

Balac

2013

IEEE Intell. Syst.

View full text Add to dashboard Cite

the grid a two-way street, with information and even power fl owing in both directions. Traditional "passive" appliances and devices are giving way to "smart" ones, which incorporate active controls and data reporting. The combination of the bidirectional power and information fl ow, when coupled with the increased local autonomy of consumer devices, has given birth to the "smart grid" concept. The Smart GridThe enabling attribute of the smart grid is the "feedback" mechanism from the consumer to the producer. The ability of the consumer to provide usage data, and in some cases even power, that's sent back to the utility enables an entirely new paradigm of power distribution and control. Smart grid goals are focused on operating efficiently and resiliently to satisfy increasing demand, while maintaining lower operating costs, reducing environmental impact, incorporating renewables, and leveraging the existing legacy grid infrastructure. The bidirectional fl ow of both electricity and information in the smart grid allows for active demand management. Feedback from end-user applications enables demand to be actively managed in real time, facilitating electricity generation at scale from intermittent renewable sources. 1 Additionally, enabling active participation by consumers in demand response has created complex dynamic environments with the potential of generating greater effi ciency and resilience, while further reducing environmental impact. Novel algorithms, methods for learning, scheduling, and optimization techniques are needed to achieve these goals. Smart MetersThe development, implementation, and integration of the new equipment, appliances, and software to execute the technologies necessary to create and support the smart grid are an ongoing process. These emerging technologies have a variety of aims, including saving energy and pursuing lowest rates, as well as contributing to the smooth and effi cient functioning of the electric grid. Computerized controls for homes and appliances can now respond in automated ways to signals from the energy provider to minimize energy cost, reduce energy consumption at times of high-demand stress, or actively shift power use to times when power is available at a lower cost. 2 The usage of "smart meter" technology is becoming more prevalent. Smart meters implement local monitoring and control functions at the end-user's location, coupled with a communication link back to the energy supplier. The information from the smart meter can be run through a home energy management system (EMS). A home EMS enables easyto-use, real-time, digital energy usage tracking and provides opportunities for energy savings. The EMS may actively affect various (smart) appliances and electronic products, as well as avoid peak demand rates, help to balance the energy load, and prevent blackouts both manually and automatically.Such efforts have enabled the concept of demandside management to evolve as an approach to actively reduce energy demand during peak hours. The goal of demand side managem...

show abstract

Self-Consumption Optimization of Renewable Energy Production in Distributed Clouds

Camus

Blavette

Dufossé

et al. 2018

2018 IEEE International Conference on Cluster Computing (CLUSTER)

View full text Add to dashboard Cite

The growing appetite of new technologies, such as Internet-of-Things, for Cloud resources leads to an unprecedented energy consumption for these infrastructures. In order to make these energy-hungry distributed systems more sustainable, Cloud providers resort more and more to on-site renewable energy production facilities like photovoltaic panels. Yet, this intermittent and variable electricity production is often uncorrelated with the Cloud consumption induced by its workload. Geographical load balancing, virtual machine (VM) migration and consolidation can be used to exploit multiple Cloud data centers' locations and their associated photovoltaic panels for increasing their renewable energy consumption. However, these techniques cost energy and network bandwidth, and this limits their utilization. In this paper, we propose to rely on the flexibility brought by Smart Grids to exchange renewable energy between distributed sites and thus, to further increase the overall Cloud's self-consumption of the locally-produced renewable energy. Our solution is named SCORPIUS: Self-Consumption Optimization of Renewable energy Production In distribUted cloudS. It takes into account telecommunication network constraints and electrical grid requirements to optimize the Cloud's self-consumption by trading-off between VM migration and renewable energy exchange. Our simulation-based results show that SCORPIUS outperforms existing solutions on various workload traces of production Clouds in terms of both renewable self-consumption and overall energy consumption.

show abstract

Energy and Thermal Aware Scheduling in Data Centers

Cited by 3 publications

References 35 publications

The Need for Traffic Based Virtualisation Management for Sustainable Clouds

The Need for Traffic Based Virtualisation Management for Sustainable Clouds

"Green Machine" Intelligence: Greening and Sustaining Smart Grids

Self-Consumption Optimization of Renewable Energy Production in Distributed Clouds

Contact Info

Product

Resources

About