Algorithms for energy conservation in heterogeneous data centers

Albers, Susanne; Quedenfeld, Jens

doi:10.1016/j.tcs.2021.10.009

Cited by 3 publications

(4 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If the operating costs are load independent, i.e., f j (z) = l j = const for all j ∈ [d], then the load-dependent operating cost L t, j (X A ) is always zero. Thus, the competitive ratio of algorithm A is 2d, so it matches the lower bound given in [5]. In contrast to the deterministic 2d-competitive online algorithm presented in [5], our algorithm can handle inefficient server types, which were excluded in [5].…”

Section: Lemma 4 For All T ∈ [T ] and J ∈supporting

confidence: 61%

“…Thus, the competitive ratio of algorithm A is 2d, so it matches the lower bound given in [5]. In contrast to the deterministic 2d-competitive online algorithm presented in [5], our algorithm can handle inefficient server types, which were excluded in [5].…”

Section: Lemma 4 For All T ∈ [T ] and J ∈supporting

confidence: 61%

“…In [5], we analyzed the discrete setting for heterogeneous data centers where the operating cost does neither depend on the load nor on time , i.e., f t, j (z) = l j = const. In this case, the total operating cost at time t is given by g t (x 1 , .…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Algorithms for Right-Sizing Heterogeneous Data Centers

Albers¹,

Quedenfeld²

2021

Preprint

Self Cite

View full text Add to dashboard Cite

Power consumption is a dominant and still growing cost factor in data centers. In time periods with low load, the energy consumption can be reduced by powering down unused servers. We resort to a model introduced by Lin, Wierman, Andrew and Thereska [23,24] that considers data centers with identical machines, and generalize it to heterogeneous data centers with d different server types. The operating cost of a server depends on its load and is modeled by an increasing, convex function for each server type. In contrast to earlier work, we consider the discrete setting, where the number of active servers must be integral. Thereby, we seek truly feasible solutions. For homogeneous data centers (d = 1), both the offline and the online problem were solved optimally in [3,4].In this paper, we study heterogeneous data centers with general time-dependent operating cost functions. We develop an online algorithm based on a work function approach which achieves a competitive ratio of 2d + 1 + ǫ for any ǫ > 0. For time-independent operating cost functions, the competitive ratio can be reduced to 2d + 1. There is a lower bound of 2d shown in [5], so our algorithm is nearly optimal. For the offline version, we give a graph-based (1 + ǫ)-approximation algorithm. Additionally, our offline algorithm is able to handle time-variable data-center sizes.

show abstract

Section: Lemma 4 For All T ∈ [T ] and J ∈supporting

confidence: 61%

Section: Lemma 4 For All T ∈ [T ] and J ∈supporting

confidence: 61%

See 1 more Smart Citation

Algorithms for Right-Sizing Heterogeneous Data Centers

Albers¹,

Quedenfeld²

2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Deep learning, as one of the key technologies for the explosion of arti cial intelligence, has made breakthroughs in the elds of computer vision and natural language processing, but the speci c applications in economics and nance still require more research [10,11]. Albers and Quedenfeld (2021) [12] pointed out that power consumption is a major cost factor for data centers, which can be reduced by dynamically adjusting the data center's size according to the currently arriving jobs. Suppose the load is low for a long time and power off the server to save energy.…”

mentioning

confidence: 99%

The Blockchain Technology Applied in the Development of Real Economy in Jiangsu under Deep Learning

Shi

Huang²

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

This study focuses on the financing difficulties of small and medium enterprises (SMEs) in China to study the application of blockchain technology in developing the real economy. Deep learning neural network is applied to the vulnerability analysis and detection of smart contracts in blockchain technology by analyzing the connotation of blockchain technology and deep learning. A multiparty joint financial service platform based on blockchain technology is established to help SMEs financing institutions reduce transaction costs, thereby helping them reduce loan interest rates. Finally, Jiangsu Province is studied as a pilot unit. The results show that the Recall and F-score of Bidirectional Neural Network for smart contract vulnerability detection are higher than those of the original neural network. The Recall rate and F-score value of the Wide and Deep model are up to 96.2% and 94.7%, which are higher than those of other vulnerability detection schemes. The Timestamp vulnerability has the highest Recall rate, 94.2%, which can rely on a large amount of valid data to improve detection efficiency. The distribution of financing needs of SMEs in Jiangsu Province from 2020 to 2021 shows that the loan number of SMEs is generally not high. Still, financial institutions and enterprises must spend the same transaction cost. After a technology company in Nanjing made a loan through a blockchain financial service platform, its financing cost decreased by 0.5331%. Blockchain technology has played a great role in the financing process of SMEs, reducing intermediate links and credit costs, and promoting the development of SMEs and the real economy.

show abstract

Transmission Failure Analysis of Multi-Protection Routing in Data Center Networks With Heterogeneous Edge-Core Servers

Wanling

Chang

et al. 2022

IEEE/ACM Trans. Networking

View full text Add to dashboard Cite

Algorithms for energy conservation in heterogeneous data centers

Cited by 3 publications

References 24 publications

Algorithms for Right-Sizing Heterogeneous Data Centers

Algorithms for Right-Sizing Heterogeneous Data Centers

The Blockchain Technology Applied in the Development of Real Economy in Jiangsu under Deep Learning

Transmission Failure Analysis of Multi-Protection Routing in Data Center Networks With Heterogeneous Edge-Core Servers

Contact Info

Product

Resources

About