Energy-Aware Data Transfer Tuning

Alan, Ismail; Arslan, Engin; Kosar, Tevfik

doi:10.1109/ccgrid.2014.117

Cited by 24 publications

(13 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is due the fact that data transfer servers on XSEDE have four cores and energy consumption per core decreases as the number of active cores increases [4]. When concurrency level goes above 4, then cores start running more data transfer threads which leads to increase in energy consumption per core.…”

Section: Resultsmentioning

confidence: 97%

“…By setting pipelining to relatively high values, we are transferring multiple data packets back-to-back that prevents idleness of the network and system resources, which in turn decreases the energy consumption. Moreover, we assigned most of the available data channels to the Small chunk which multiplies the impact of energy saving when combined with pipelining [4]. For the parallelism level, we again consider TCP buffer size, BDP, and average file size.…”

Section: Minimum Energy Transfer Algorithmmentioning

confidence: 99%

“…In another study [10], we presented four application level dynamic data transfer algorithms that tune the transfer parameters heuristically for reaching optimum network throughput but we did not consider the power consumption of these algorithms. In [4], we analyzed how each data transfer parameter may affect both the throughput and the power consumption in the end-toend data transfers.…”

Section: Related Workmentioning

confidence: 99%

“…There has been studies on Energy Efficient Ethernet [3] for making ethernet cards more energy efficient, but these were very limited in the scope since they were only considering the energy saving of the ethernet card and not the entire endsystem (including the CPU, memory, disk etc). Prior studies [4,28,34] show that at least one quarter of the data transfer power consumption happens at the end-systems (sender and receiver nodes) and the rest at the networking infrastructure. Significant amount of data transfer energy savings can be obtained at the end-systems with no or minimal performance penalty.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Energy-aware data transfer algorithms

Alan

Arslan

Kosar

2015

Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis

Self Cite

View full text Add to dashboard Cite

The amount of data moved over the Internet per year has already exceeded the Exabyte scale and soon will hit the Zettabyte range. To support this massive amount of data movement across the globe, the networking infrastructure as well as the source and destination nodes consume immense amount of electric power, with an estimated cost measured in billions of dollars. Although considerable amount of research has been done on power management techniques for the networking infrastructure, there has not been much prior work focusing on energy-aware data transfer algorithms for minimizing the power consumed at the end-systems. We introduce novel data transfer algorithms which aim to achieve high data transfer throughput while keeping the energy consumption during the transfers at the minimal levels. Our experimental results show that our energy-aware data transfer algorithms can achieve up to 50% energy savings with the same or higher level of data transfer throughput.

show abstract

Section: Resultsmentioning

confidence: 97%

Section: Minimum Energy Transfer Algorithmmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Energy-aware data transfer algorithms

Alan

Arslan

Kosar

2015

Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis

Self Cite

View full text Add to dashboard Cite

show abstract

“…Considering more components of a server, Tudor et al [14] propose a power model expressed as a function of energy used by CPU, memory, and I/O devices. Furthermore, Song et al [15] describe a similar power model by expressing the system power consumption as a summation of CPU, memory, disk, and network interface card, which can be shown as

E_{total} = E_{cpu} + E_{memory} + E_{disk} + E_{NIC}

Another energy model can be further constructed considering the levels of resource utilisation by the key components of a server as [16]

P_{t} = C_{cpu} U_{cpu} + C_{memory} U_{memory} + C_{disk} U_{disk}

where

U_{cpu}

is the CPU utilisation,

U_{memory}

is the memory access rate,

U_{disk}

is the hard disk I/O request rate, and

U_{NIC}

is the network I/O request rate.

P_{t}

refers to the predicted power consumption of server at time t while

C_{cpu}

C_{memory}

C_{disk}

, and

C_{NIC}

are the coefficients of CPU, memory, disk, and NIC, respectively.…”

Section: Related Workmentioning

confidence: 99%

Power consumption model based on feature selection and deep learning in cloud computing scenarios

Yang

2020

IET Communications

View full text Add to dashboard Cite

EWM: An entropy‐based framework for estimating energy consumption of edge servers

Li,

2023

Engineering Reports

View full text Add to dashboard Cite

In mobile edge computing (MEC), accurately predicting and monitoring the energy consumption of edge servers is a key challenge in achieving green computing. The importance of solving this problem is that it can help optimize the energy usage in data centers and thus reduce the carbon emission of MEC. To this end, we propose an innovative entropy‐based power modeling framework called entropy weighted model (EWM). The EWM framework weights and combines classical prediction models by analyzing the major components of a server and selecting appropriate parameters. We validate the performance of EWM using real server power and performance counter data and compare it with other classical prediction models by Friedman test. The results show that EWM outperforms other classical prediction models in all test datasets. This result validates the significant advantages of our EWM framework in solving the critical problem of edge server power prediction, and provides an effective tool for achieving data center energy optimization and promoting green computing, resulting in a highly general and accurate prediction model.

show abstract

Energy-Aware Data Transfer Tuning

Cited by 24 publications

References 25 publications

Energy-aware data transfer algorithms

Energy-aware data transfer algorithms

Power consumption model based on feature selection and deep learning in cloud computing scenarios

EWM: An entropy‐based framework for estimating energy consumption of edge servers

Contact Info

Product

Resources

About