On energy-aware aggregation of dynamic temporal demand in cloud computing

Qian, Haiyang; Li, Fu; Medhi, Deep

doi:10.1109/comsnets.2012.6151370

Cited by 9 publications

(7 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We proposed models to minimize server costs in data centers and introduce workload aggregation methods that combat the time complexity to compute the optimal cost in large scale data centers in our earlier workshop paper [36]. This paper is an extension of this early work.…”

Section: Insight-3mentioning

confidence: 95%

See 1 more Smart Citation

Optimal Resource Provisioning and the Impact of Energy-Aware Load Aggregation for Dynamic Temporal Workloads in Data Centers

Qian

Li²,

Ravindran

et al. 2014

IEEE Trans. Netw. Serv. Manage.

Self Cite

View full text Add to dashboard Cite

An important goal of data center providers is to minimize their operational cost, which reflected through the wear-and-tear cost and the energy consumption cost. In this paper, we present optimization formulations to minimize the cost of ownership in terms of server energy consumption and serverwear-and-tear cost under three different data center server setups (homogeneous, heterogeneous, and hybrid hetero-homogeneous clusters) for dynamic temporal workloads. Our studies show that the homogeneous model takes significantly less computational time than the heterogeneous model (by an order of magnitude). To compute optimal configurations in near real time for large-scale data centers, we propose two modes for using our models: aggregation by maximum (preserves workload deadline) and aggregation by mean (relaxes workload deadline). In addition, we propose two aggregation methods for use in each of the two modes: static (periodic) aggregation and dynamic (aperiodic)aggregation. We found that in the aggregation by maximum mode, dynamic aggregation resulted in cost savings of up to approximately 18% over the static aggregation. In the aggregation by mean mode, dynamic aggregation saved up to approximately a 50% workload rearrangement compared with the static aggregationby mean mode.

show abstract

Section: Insight-3mentioning

confidence: 95%

“…This paper is an extension of this early work. Compared to [36], there are several major improvements. First, the study conducted is comprehensive and is scaled up to solve large-scale problems; we also considered the consolidation cost when switching off machines in this study.…”

Section: Insight-3mentioning

confidence: 95%

Optimal Resource Provisioning and the Impact of Energy-Aware Load Aggregation for Dynamic Temporal Workloads in Data Centers

Qian

Li²,

Ravindran

et al. 2014

IEEE Trans. Netw. Serv. Manage.

Self Cite

View full text Add to dashboard Cite

show abstract

“…In this case, we consider applying sampling techniques [26][27][28][29][30][31][32][33][34] to perform fast and accurate estimation for it. Last but not least, we are interested in applying our digraph spectral graph method to detect node and link failures in large-scale cloud computing environments [35][36][37][38].…”

Section: Resultsmentioning

confidence: 99%

Digraph Spectral Clustering with Applications in Distributed Sensor Validation

Zhai

2014

International Journal of Distributed Sensor Networks

View full text Add to dashboard Cite

In various sensor networks, the performances of sensors vary significantly over time, due to the changes of surrounding environment, device hardware, and so forth. Hence, monitoring the status is essential in sensor network maintenance. Spectral clustering has been employed as an enabling technique to solve this problem. However, the traditional spectral clustering is developed for undirected graph, and the naive generalization for directed graph by symmetrization of the adjacency matrix will lead to loss of network information, and thus cannot efficiently detect bad sensor nodes while applying it for sensor validation. In this paper, we develop a generalized digraph spectral clustering method. Instead of simply symmetrizing the adjacency matrix, our method takes into consideration the network circulation while clustering the sensors. The extensive simulation results demonstrate that our method outperforms the traditional spectral clustering method by increasing the bad detection ratio from 19% to 41%.

show abstract

“…However, migration has a number of energy cost components. If Eof f h1 is the energy used to turn off the source server of migration, Eon h2 is the energy used to turn on the destination server of migration (if previously turned off), Enm is the energy consumed by the network during the migration process, Et h1 is the energy consumed during the time of the migration by the source server, and Et h2 is the energy consumed during the time of the migration by the destination server, then the total energy cost Ec of process migration is given by (adapted from [9]):…”

Section: Network Speed and Migrationsmentioning

confidence: 99%

High speed network impacts and power consumption estimation for cloud data centers

Lago

Madeira

Medhi

2015

Proceedings of the 30th Annual ACM Symposium on Applied Computing

Self Cite

View full text Add to dashboard Cite

Networking technology in data centers is limited by Ethernet speed, and a standard for 10 Tbps Ethernet is already envisioned. In this paper, we seek to understand and analyze the energy impact of increasing the network speed in the energy consumption in data centers that operate in a cloud with virtual machine migration. We consider four scheduling algorithms of virtual machines for workload scheduling. Results obtained by simulation show that from the energy consumption point of view, the bandwidth of non-federated data centers significantly impacts the intra cloud scheduling process, and this should be considered when choosing a scheduling algorithm. We also found that the makespan has minimal impact with this increased bandwidth in data centers, while the number of migrations increases greatly. We also develop an empirical model to estimate the power consumption based on bandwidth for studied scenarios, where there is an upper limit of bandwidth that impacts the energy consumption, after which virtually no changes are noticed.

show abstract

On energy-aware aggregation of dynamic temporal demand in cloud computing

Cited by 9 publications

References 7 publications

Optimal Resource Provisioning and the Impact of Energy-Aware Load Aggregation for Dynamic Temporal Workloads in Data Centers

Optimal Resource Provisioning and the Impact of Energy-Aware Load Aggregation for Dynamic Temporal Workloads in Data Centers

Digraph Spectral Clustering with Applications in Distributed Sensor Validation

High speed network impacts and power consumption estimation for cloud data centers

Contact Info

Product

Resources

About