Filters








109,309 Hits in 4.9 sec

Cluster-level feedback power control for performance optimization

Xiaorui Wang, Ming Chen
2008 High-Performance Computer Architecture  
In this paper, we propose a cluster-level power controller that shifts power among servers based on their performance needs, while controlling the total power of the cluster to be lower than a constraint  ...  As a result, at the cluster level where multiple servers are correlated by common workloads and share common power supplies, power cannot be shared to improve application performance.  ...  Acknowledgements We would like to thank Charles Lefurgy and Malcolm Ware at the IBM Austin Research Laboratory for their insightful discussions, and anonymous reviewers for their valuable comments.  ... 
doi:10.1109/hpca.2008.4658631 dblp:conf/hpca/WangC08 fatcat:aososfuyarhjheb5y2npku3hou

Power and QoS Aware Multi-level Resource Coordination and Scheduling in Virtualized Servers

Congfeng Jiang, Jingling Mao, Dongyang Ou, Yumei Wang, Xindong You, Jilin Zhang, Jian Wan
2016 International Journal of Grid and Distributed Computing  
This three-level controlling framework schedules and allocates hardware resource for QoS guarantee and cluster level power management.  ...  Cluster level power and QoS coordination is crucial for data center level energy efficiency coordination as well as high quality service provisioning.  ...  They only analyze the cluster power and performance from two tiers while our three-level controlling framework schedules and allocates hardware resource for QoS guarantee and cluster level power management  ... 
doi:10.14257/ijgdc.2016.9.11.27 fatcat:bcrprhh6ffby7cvw72kdyklroq

Cluster-based feedback control of turbulent post-stall separated flows [article]

Aditya G. Nair, Chi-An Yeh, Eurika Kaiser, Bernd R. Noack, Steven L. Brunton, Kunihiko Taira
2018 arXiv   pre-print
A feedback control law is then sought for each cluster state through iterative evaluation and downhill simplex search to minimize power consumption in flight.  ...  We propose a novel model-free self-learning cluster-based control strategy for general nonlinear feedback flow control technique, benchmarked for high-fidelity simulations of post-stall separated flows  ...  For the optimized feedback control law for 2D separated flow, suction is performed for the low-drag clusters C D /C D < 1, while blowing is performed for the remaining clusters.  ... 
arXiv:1809.07220v1 fatcat:6we5veo47fe3fns3neyx226cfq

Adaptive power control for server clusters

Xiaorui Wang, Ming Chen
2008 Proceedings, International Parallel and Distributed Processing Symposium (IPDPS)  
In this paper, we propose a cluster-level power controller that shifts power among servers based on their performance needs, while controlling the total power of the cluster to be lower than a constraint  ...  As a result, at the cluster level where multiple servers are correlated by common workloads and share common power supplies, power cannot be shared to improve application performance.  ...  ACKNOWLEDGEMENTS We would like to thank Charles Lefurgy and Malcolm Ware at the IBM Austin Research Laboratory for their insightful discussions. This work is funded in part by NSF under Grant No.  ... 
doi:10.1109/ipdps.2008.4536425 dblp:conf/ipps/WangC08 fatcat:xtzot37klvbzndklbcwxtwajhi

Damping of Inter-Area Oscillations via Modulation of Aggregated Loads

Felipe Wilches-Bernal, Raymond H. Byrne, Jianming Lian
2019 IEEE Transactions on Power Systems  
The load cluster control action is obtained from an optimal output feedback control (OOFC) strategy.  ...  The simulation results show that the proposed controllers successfully damp inter-area oscillations under different operating conditions and with different clustering for the events considered.  ...  Fig. 4 shows the power levels of the loads in Cluster 2 when this cluster is being used to perform the probing.  ... 
doi:10.1109/tpwrs.2019.2948116 fatcat:fmagoamxtrgjrd6d7mng4g4vwm

Robust control-theoretic thermal balancing for server clusters

Yong Fu, Chenyang Lu, Hongan Wang
2010 2010 IEEE International Symposium on Parallel & Distributed Processing (IPDPS)  
CTB features controllers rigorously designed based on optimal control theory and a difference equation model that approximates the thermal dynamics of clusters.  ...  Thermal management is critical for clusters because of the increasing power consumption of modern processors, compact server architectures and growing server density in data centers.  ...  By employing both temperature and CPU utilization feedbacks in its optimal control design, the CTB-UT algorithm provides a particularly attractive solution for server clusters, as it introduces low control  ... 
doi:10.1109/ipdps.2010.5470480 dblp:conf/ipps/FuLW10 fatcat:7pkosgm5gbacdcsov3363owoaa

Multi-objective optimization of demand response in a datacenter with lithium-ion battery storage

A. Mamun, I. Narayanan, D. Wang, A. Sivasubramaniam, H.K. Fathy
2016 Journal of Energy Storage  
Our optimization study focuses on a classical feedforward-feedback energy management policy, where feedforward control is used for peak shaving, and feedback is used for tracking a desired battery state  ...  We optimize these architectures' parameters using differential evolution, for real datacenter power demand histories.  ...  The authors would also like to thank Kushagra Vaid from Cloud Server Infrastructure team in Microsoft for providing power traces for validating the battery model.  ... 
doi:10.1016/j.est.2016.08.002 fatcat:n6prcil465h4fimjgbzchxmmui

An Efficient Threshold-Based Power Management Mechanism for Heterogeneous Soft Real-Time Clusters

Leping Wang, Ying Lu
2010 IEEE Transactions on Industrial Informatics  
With growing cost of electricity, the power management (PM) of server clusters has become an important problem.  ...  Considering the increasing popularity of heterogeneous systems, this paper proposes an efficient algorithm for PM of heterogeneous soft real-time clusters.  ...  For illustration, the curves in Fig. 11(b) show the power consumption of TP-CP-OP algorithm with and without DVS feedback control.  ... 
doi:10.1109/tii.2010.2047950 fatcat:s22mo2hu4jf5fhk3ppeegaz75i

Efficient Power Management of Heterogeneous Soft Real-Time Clusters

Leping Wang, Ying Lu
2008 2008 Real-Time Systems Symposium  
Considering the increasing popularity of heterogeneous systems, this paper proposes an efficient algorithm for power management of heterogeneous soft real-time clusters.  ...  With growing cost of electricity, the power management of server clusters has become an important problem. However, most previous researchers only address the challenge in homogeneous environments.  ...  For illustration, Figure 10 presents the power consumption curves of TP-CP-OP algorithm with and without feedback DVS control.  ... 
doi:10.1109/rtss.2008.31 dblp:conf/rtss/WangL08 fatcat:cv53lzm4xvhinbjr6zdo7fcvvm

Multi-mode energy management for multi-tier server clusters

Tibor Horvath, Kevin Skadron
2008 Proceedings of the 17th international conference on Parallel architectures and compilation techniques - PACT '08  
Moreover, we rigorously investigate selection of the optimal number of spare servers for each power state, a problem that has only been approached in an ad-hoc manner in current policies.  ...  little or no performance degradation.  ...  Transient demand variations are compensated for with PID feedback control.  ... 
doi:10.1145/1454115.1454153 dblp:conf/IEEEpact/HorvathS08 fatcat:rln7qhojsnhhnmpa5lw4hpn2xq

An integrated energy aware wireless transmission system for QoS provisioning in wireless sensor network

Yong Yuan, Zongkai Yang, Zhihai He, Jianhua He
2006 Computer Communications  
heads including the modulation level (bits per symbol) and transmit power.  ...  In this framework, the cluster-based protocol is used, where the cluster heads form a multi-hop backbone. The intra-cluster QoS performance and inter-cluster QoS performance are both analyzed.  ...  Acknowledgements The authors would like to thank the editors and anonymous reviewers for their valuable comments.  ... 
doi:10.1016/j.comcom.2005.05.015 fatcat:wjkbyvx5bzhpxn3f63ofxp2mxe

Machine Learning in Action: Examples [chapter]

Mariette Awad, Rahul Khanna
2015 Efficient Learning Machines  
The learning process constructs a knowledge base that is hardened by critical feedback to improve performance.  ...  These examples demonstrate an intelligent feedback control system based on the principles of machine learning.  ...  Learning Process: Monitoring Achieving power-efficient performance and abiding by power and performance constraints call for real-time feedback control.  ... 
doi:10.1007/978-1-4302-5990-9_11 fatcat:mdmgulxgljginjs7jdwjdxg7qe

A Survey of Power-Saving Techniques on Data Centers and Content Delivery Networks

Chang Ge, Zhili Sun, Ning Wang
2013 IEEE Communications Surveys and Tutorials  
Furthermore, we present a high-level overview of the literature by categorizing existing approaches with respect to their scopes and research directions.  ...  efforts have also been devoted to the design of practical powersaving techniques in content delivery networks (CDNs), as they involve thousands of globally distributed data centers with content server clusters  ...  Firstly, a power controller operates at cluster level to perform dynamic cluster-level power capping and server-level power shifting.  ... 
doi:10.1109/surv.2012.102512.00019 fatcat:nw5r35dh4bh3rdtp62xgyaa2ei

A Frequency Control Strategy Considering Large Scale Wind Power Cluster Integration Based on Distributed Model Predictive Control

Bohao Sun, Yong Tang, Lin Ye, Chaoyu Chen, Cihang Zhang, Wuzhi Zhong
2018 Energies  
and single turbine level, which implements multi-objectives coordination optimization control.  ...  In the active power dispatch aspect, Ye et al. [24] divided wind turbines into groups according to the generation status, and rolling optimization is utilized at the wind farm level, wind turbines level  ...  At the wind farm and wind power cluster level, there are many studies on hierarchical control frameworks for active power dispatch [17, 18] .  ... 
doi:10.3390/en11061600 fatcat:mnkfj6aygfgdbbzrbsexbbkvjq

Provisioning Statistical QoS for Coordinated Communications with Limited Feedback

Mohammad G. Khoshkholgh, Keivan Navaie, Kang G. Shin, Victor C. M. Leung
2016 2016 IEEE Global Communications Conference (GLOBECOM)  
Of crucial practical importance, the findings of this paper also indicates that in CoMP there is an optimal cluster size for a given feedback capacity that maximizes the corresponding EC.  ...  This paper considers the optimal partitioning of the feedback bits in CoMP while accounting for the inter-cell interference cancellation (ICIC).  ...  and it should be joined with other techniques such as power control and dynamic clustering, (iii) power control is highly effective in conjunction with optimal FBP; otherwise the performance boost is  ... 
doi:10.1109/glocom.2016.7842179 dblp:conf/globecom/KhoshkholghNSL16 fatcat:5wimdkgi5bhrni3vqnwp7bvwmm
« Previous Showing results 1 — 15 out of 109,309 results