Survey of Techniques and Architectures for Designing Energy-Efficient Data Centers

Shuja, Junaid; Bilal, Kashif; Madani, Sajjad A.; Othman, Mazliza; Ranjan, Rajiv; Balaji, Pavan; Khan, Samee U.

doi:10.1109/jsyst.2014.2315823

Cited by 157 publications

(85 citation statements)

References 90 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This model is proposed to reduce the response time of requests by reducing the delay time that comes from the distance between the computing resources and the end users and the queuing time inside the computing resources themselves [5]. The computing resources may refer to cloudlets, edge servers, micro-datacenters, or fog nodes [6,7].…”

Section: Related Workmentioning

confidence: 99%

Delay-aware power optimization model for mobile edge computing systems

Jararweh

Al‐Ayyoub

Al-Quraan

et al. 2017

Pers Ubiquit Comput

View full text Add to dashboard Cite

Reducing the total power consumption and network delay are among the most interesting issues facing large-scale Mobile Cloud Computing (MCC) systems and their ability to satisfy the Service Level Agreement (SLA). Such systems utilize cloud computing infrastructure to support offloading some of user's computationally heavy tasks to the cloud's datacenters. However, the delay incurred by such offloading process lead the use of servers (called cloudlets) placed in the physical proximity of the users, creating what is known as Mobile Edge Computing (MEC). The cloudlet-based infrastructure has its challenges such as the limited capabilities of the cloudlet system (in terms of the ability to serve different request types from users in vast geographical regions). To cover the users demand for different types of services and in vast geographical regions, cloudlets cooperate among each other by passing user requests from one cloudlet to another. This cooperation affects both power consumption and delay. In this work, we present a mixed integer linear programming (MILP) optimization model for MEC systems with these two issues in mind. Specifically, we consider two types of cloudlets: local cloudlets and global cloudlets, which have higher capabilities. A user connects to a local cloudlet and sends all of its traffics to it. If the local cloudlet cannot serve the desired request, then the request is moved to another local cloudlet. If no local cloudlet can serve the request, then it is moved to a global cloudlet which can serve all service types. The process of routing requests through the hierarchical network of cloudlets increases power consumption and delay. Our model minimizes power consumption while incurring an acceptable amount of delay. We evaluate it under several realistic scenarios to show that it can indeed be used for power optimization of large-scale MEC systems without violating delay constraints.

show abstract

Section: Related Workmentioning

confidence: 99%

Delay-aware power optimization model for mobile edge computing systems

Jararweh

Al‐Ayyoub

Al-Quraan

et al. 2017

Pers Ubiquit Comput

View full text Add to dashboard Cite

show abstract

“…on the way to smart world, the energy efficiency for DCs becomes more pressing [5]. About green DC, possible techniques to improve energy efficiency can be achieved from the following aspects [10], [22], [30]. 1) Use renewable or green sources of energy (e.g., wind, water, solar energy, heat pumps, etc.…”

Section: Green Data Center (Dc)mentioning

confidence: 99%

Green IoT Agriculture and HealthcareApplication (GAHA)

Nandyala¹,

Kim²

2016

IJSH

View full text Add to dashboard Cite

show abstract

“…On the other hand, the near infinite resources of cloud computing present an ideal candidate for computational offloading. Cloud Data Centers (CDCs) are put together with thousands of compute and store nodes that are connected with each other and the userfacing web interfaces [3]. CDCs provide a flexible pricing and utilization model over the internet that can utilized by resource constrained mobile devices.…”

Section: Introductionmentioning

confidence: 99%