Carbon-Aware Energy Capacity Planning for Datacenters

Ren, Chuangang; Wang, Di; Urgaonkar, Bhuvan; Sivasubramaniam, Anand

doi:10.1109/mascots.2012.51

Cited by 135 publications

(88 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The most common energy sources of carbon emission rates are shown in [3,4]. Here, we note that renewable energy sources have a significantly lower carbon emission rate than fossil fuels such as oil, gas, and coal.…”

Section: Introductionmentioning

confidence: 83%

“…According to current trend estimates, the United States' data centers alone are expected to consume about 73 billion kWh per year by 2020 [2]. In fact, all the data centers around the world devour more power than most countries in the world, except for four countries [3]. However, this massive energy consumption results not only in high electricity costs, but also in high carbon emissions.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

EH-GC: An Efficient and Secure Architecture of Energy Harvesting Green Cloud Infrastructure

et al. 2017

View full text Add to dashboard Cite

Nowadays, the high power consumption of data centers is the biggest challenge to making cloud computing greener. Many researchers are still seeking effective solutions to reduce or harvest the energy produced at data centers. To address this challenge, we propose a green cloud infrastructure which provides security and efficiency based on energy harvesting (EH-GC). The EH-GC is basically focused on harvesting the heat energy produced by data centers in the Infrastructure-as-a-Service (IaaS) infrastructure. A pyroelectric material is used to generate the electric current from heat using the Olsen cycle. In order to achieve efficient green cloud computing, the architecture utilizes a genetic algorithm for proper virtual machine allocation, taking into consideration less Service Level Agreement (SLA) violations. The architecture utilizes Multivariate Correlation Analysis (MCA) correlation analysis based on a triangular map area generation to detect Denial of Service (DoS) attacks in the data center layer of the IaaS. Finally, the experimental analysis is explained based on the energy parameter, which proves that our model is efficient and secure, and that it efficiently reuses the energy emitted from the data center.

show abstract

Section: Introductionmentioning

confidence: 83%

Section: Introductionmentioning

confidence: 99%

EH-GC: An Efficient and Secure Architecture of Energy Harvesting Green Cloud Infrastructure

et al. 2017

View full text Add to dashboard Cite

show abstract

“…A data center can reduce its carbon footprint by purchasing RECs [28,61]. Similarly, a data center that generates renewable energy on-site can offer its spare green energy in the market by issuing RECs, allowing other data centers to offset their carbon impact [61].…”

Section: Exploit Opportunities In the Energy Marketsmentioning

confidence: 99%

Toward sustainable data centers: a comprehensive energy management strategy

Guitart

2016

Computing

View full text Add to dashboard Cite

Data centers are major contributors to the emission of carbon dioxide to the atmosphere, and this contribution is expected to increase in the following years. This has encouraged the development of techniques to reduce the energy consumption and the environmental footprint of data centers. Whereas some of these techniques have succeeded to reduce the energy consumption of the hardware equipment of data centers (including IT, cooling, and power supply systems), we claim that sustainable data centers will be only possible if the problem is faced by means of a holistic approach that includes not only the aforementioned techniques but also intelligent and unifying solutions that enable a synergistic and energy-aware management of data centers. In this paper, we propose a comprehensive strategy to reduce the carbon footprint of data centers that uses the energy as a driver of their management procedures. In addition, we present a holistic management architecture for sustainable data centers that implements the aforementioned strategy, and we propose design guidelines to accomplish each step of the proposed strategy, referring to related achievements and enumerating the main challenges that must be still solved.

show abstract

“…For longer expected absorption times the exact method based on the matrix inversion would not be feasible any more, due to the numerical instability of the matrix inversion required by Equation (12) and (13). The figures clearly show that, as the inter-arrival time increases, making the mean absorption time longer, the relative error of the approximation rapidly decreases.…”

Section: Validation Of the Approximationmentioning

confidence: 99%

“…Among these we mention the energy consumption, which is one of the important costs that a cloud service supplier has to support. This is witnessed by the many efforts that the research community has devoted to the definition of administration policies which tend to reduce the energy consumption while maintaining reasonable levels for the Quality of Service (QoS), see, e.g., [8,13,9] just to mention a non exhaustive list of recent works. Informally, the idea behind these works is that the computational power of the cloud infrastructure is reduced when the workload is light so that the overall power consumption is reduced.…”

Section: Introductionmentioning

confidence: 99%

Evaluating the impact of eDoS attacks to cloud facilities

Rossi¹,

Iacono²,

Marin³

2016

Proceedings of the 9th EAI International Conference on Performance Evaluation Methodologies and Tools

View full text Add to dashboard Cite

The complexity of modern cloud facilities requires attentive management policies that should encompass all aspects of the system. Security is a critical issue, as intrusions, misuse or denial of service attacks may damage both the users and the cloud provider including its reputation on the market. Disruptive attacks happen fast, cause evident and short term damages and are usually the result of operations that are hard to disguise. On the other hand, Energy oriented Denial of Service (eDoS) attacks aim at producing continuous minor damages, eventually with long term consequences. These long lasting attacks are difficult to detect. In this paper we model and analyse the behaviour of a system under eDoS attack. We study the impact in terms of cloud energy consumption of an attack strategy previously proposed in the literature and compare it with other strategies that we propose. Our findings show that the strategy previously proposed in the literature, based on keeping the cloud close to saturation, is not optimal (from the point of view of the attacker) in presence of non-constant workload and that there is a trade-off between the aggressiveness of the attacker and the duration of the attack in order to maximise the damage.

show abstract

Carbon-Aware Energy Capacity Planning for Datacenters

Cited by 135 publications

References 11 publications

EH-GC: An Efficient and Secure Architecture of Energy Harvesting Green Cloud Infrastructure

EH-GC: An Efficient and Secure Architecture of Energy Harvesting Green Cloud Infrastructure

Toward sustainable data centers: a comprehensive energy management strategy

Evaluating the impact of eDoS attacks to cloud facilities

Contact Info

Product

Resources

About