DENS: data center energy-efficient network-aware scheduling

Abstract

Abstract-In modern data centers, energy consumption accounts for a considerably large slice of operational expenses. The state of the art in data center energy optimization is focusing only on job distribution between computing servers based on workload or thermal profiles. This paper underlines the role of communication fabric in data center energy consumption and presents a scheduling approach that combines energy efficiency and network awareness, termed DENS. The DENS methodology balances the energy consumption of a data center, individual job performance, and traffic demands. The proposed approach optimizes the tradeoff between job consolidation (to minimize the amount of computing servers) and distribution of traffic patterns (to avoid hotspots in the data center network).

Energy-efficient data replication in cloud computing datacenters

“…It is based on Ns2 simulation platform [17] and allows capturing realistic TCP/IP behavior in a large variety of network scenarios. GreenCloud offers a detailed fine-grained modeling of the energy consumed by the data center hardware, such as servers, switches, and communication links, and implements a number of networkaware resource allocation and scheduling solutions [26], [21]. The performance evaluation considers only single datacenter with a holistic model for resource representation [9].…”

Problem Based Learning on Cloud Economics Analysis Using Open Source Simulation

“…The idea behind DNS scheme is that machines or servers in idle condition still operating and consume up to 66% of the full energy which is not economically efficient [11]. Therefore, Dynamics Shutdown scheme works by shutting down any servers and switches in idle condition after some period of time.…”

Energy Consumption Optimization in Cloud Data Centers

“…In [50], the authors developed a strategy for traditional three-tier data center architectures which takes into consideration the energy consumption of both servers and network switches. The proposed strategy analyzes the load of each network switch to avoid overloading them.…”