© 2013 The SmartenIT Consortium 1 Commercial in Confidence Game Theoretic approach to energy efficiency Mateusz Wielgosz, Krzysztof Wajda, AGH Krakow Meeting,

Slides:



Advertisements
Similar presentations
A 2030 framework for climate and energy policies Marten Westrup
Advertisements

1/17/20141 Leveraging Cloudbursting To Drive Down IT Costs Eric Burgener Senior Vice President, Product Marketing March 9, 2010.
An Improved TCP for transaction communications on Sensor Networks Tao Yu Tsinghua University 2/8/
Network Resource Broker for IPTV in Cloud Computing Lei Liang, Dan He University of Surrey, UK OGF 27, G2C Workshop 15 Oct 2009 Banff,
Improving Datacenter Performance and Robustness with Multipath TCP
VARUN GUPTA Carnegie Mellon University 1 Partly based on joint work with: Anshul Gandhi Mor Harchol-Balter Mike Kozuch (CMU) (CMU) (Intel Research)
All Rights Reserved © Alcatel-Lucent 2009 Enhancing Dynamic Cloud-based Services using Network Virtualization F. Hao, T.V. Lakshman, Sarit Mukherjee, H.
Scaling The Edge Bridge Address Table In Datacenter Networks June-2012.
Asaf Cidon. , Tomer M. London
Energy-efficient Task Scheduling in Heterogeneous Environment 2013/10/25.
2  Industry trends and challenges  Windows Server 2012: Beyond virtualization  Complete virtualization platform  Improved scalability and performance.
Resource Shares Dynamic resource management
Dynamic Traffic Management (DTM) for minimization of inter- domain traffic cost Rafal Stankiewicz, Zbigniew Dulinski AGH University of Science.
University of Minnesota Optimizing MapReduce Provisioning in the Cloud Michael Cardosa, Aameek Singh†, Himabindu Pucha†, Abhishek Chandra
1/16 Distributed Systems Architecture Research Group Universidad Complutense de Madrid An Introduction to Virtualization and Cloud Technologies to Support.
1 Dell World 2014 Dell & Trend Micro Boost VM Density with AV Designed for VDI TJ Lamphier, Sr. Director Trend Micro & Aaron Brace, Solution Architect.
Peter Key, Laurent Massoulie, Don Towsley Infocom 07 presented by Park HoSung 1 Path selection and multipath congestion control.
Scalable Rule Management for Data Centers Masoud Moshref, Minlan Yu, Abhishek Sharma, Ramesh Govindan 4/3/2013.
Cloud Computing at GES DISC Presented by: Long Pham Contributors: Aijun Chen, Bruce Vollmer, Ed Esfandiari and Mike Theobald GES DISC UWG May 11, 2011.
The Case for Enterprise Ready Virtual Private Clouds Timothy Wood, Alexandre Gerber *, K.K. Ramakrishnan *, Jacobus van der Merwe *, and Prashant Shenoy.
Walter Binder University of Lugano, Switzerland Niranjan Suri IHMC, Florida, USA Green Computing: Energy Consumption Optimized Service Hosting.
Chapter 4 Infrastructure as a Service (IaaS)
Dave Bradley Rick Harper Steve Hunter 4/28/2003 CoolRunnings.
Power Management in Cloud Computing using Green Algorithm -Kushal Mehta COP 6087 University of Central Florida.
Energy-efficient Virtual Machine Provision Algorithms for Cloud System Ching-Chi Lin Institute of Information Science, Academia Sinica Department of Computer.
Efficient Resource Management for Cloud Computing Environments Andrew J. Younge, Gregor von Laszewski, Lizhe Wang, Sonia Lopez-Alarcon, Warren Carithers.
Overcoming the challenge of virtual blindness Colin Richardson on365 Ltd.
Green Cloud Computing Hadi Salimi Distributed Systems Lab, School of Computer Engineering, Iran University of Science and Technology,
Xavier León PhD defense
SLA-aware Virtual Resource Management for Cloud Infrastructures
IT:Network:Applications VIRTUAL DESKTOP INFRASTRUCTURE.
DESIGN CONSIDERATIONS OF A GEOGRAPHICALLY DISTRIBUTED IAAS CLOUD ARCHITECTURE CS 595 LECTURE 10 3/20/2015.
Present By : Bahar Fatholapour M.Sc. Student in Information Technology Mazandaran University of Science and Technology Supervisor:
Online Auctions in IaaS Clouds: Welfare and Profit Maximization with Server Costs Xiaoxi Zhang 1, Zhiyi Huang 1, Chuan Wu 1, Zongpeng Li 2, Francis C.M.
Authors: Thomas Ristenpart, et at.
WHAT IS PRIVATE CLOUD? Michał Jędrzejczak Główny Architekt Rozwiązań Infrastruktury IT
A User Experience-based Cloud Service Redeployment Mechanism KANG Yu.
Energy Efficiency in Cloud Data Centers: Energy Efficient VM Placement for Cloud Data Centers Doctoral Student : Chaima Ghribi Advisor : Djamal Zeghlache.
MOBILE CLOUD COMPUTING
Department of Computer Science Engineering SRM University
Dynamic Resource Allocation Using Virtual Machines for Cloud Computing Environment.
Virtualization. Virtualization  In computing, virtualization is a broad term that refers to the abstraction of computer resources  It is "a technique.
MS Tech-Ed 2006 Iron Architect Competition Greg Cogdell Milliken & Co.
Copyright © 2011 EMC Corporation. All Rights Reserved. MODULE – 6 VIRTUALIZED DATA CENTER – DESKTOP AND APPLICATION 1.
Cloud Computing Energy efficient cloud computing Keke Chen.
© 2014 The SmartenIT Consortium 1 Commercial in Confidence Panel on “Cloud Federations and SDN/NFV: the highways towards improved QoE, Cost, and Energy.
Copyright 2010 – Johnson Controls, Inc. 1 A Day in the Life of a Smart Campus Clay Nesler VP, Global Energy & Sustainability Johnson Controls
© 2013 The SmartenIT Consortium 1 Commercial in Confidence 2.4: Game Theoretic approach to Horst-RB Mateusz Wielgosz, AGH ? ?, ? ?, 2014 Socially-aware.
Challenges towards Elastic Power Management in Internet Data Center.
McGraw-Hill/Irwin Copyright © 2013 by The McGraw-Hill Companies, Inc. All rights reserved. Business Plug-In B17 Organizational Architecture Trends.
Power-Aware Scheduling of Virtual Machines in DVFS-enabled Clusters
Brussels Workshop Use case 3 11/09/2015 Mario Sisinni.
Copyright © 2011, Performance Evaluation of a Green Scheduling Algorithm for Energy Savings in Cloud Computing Truong Vinh Truong Duy; Sato,
Haley Toni Mastelić & Dražen Lučanin, Vienna University of Technology.
VGreen: A System for Energy Efficient Manager in Virtualized Environments G. Dhiman, G Marchetti, T Rosing ISLPED 2009.
FYP Briefing Presentation Building an Efficient IaaS: - Let’s become experts in cloud computing! April 15, 2010.
Copyright © 2010, Performance and Power Management for Cloud Infrastructures Hien Nguyen Van; Tran, F.D.; Menaud, J.-M. Cloud Computing (CLOUD),
Cloud Computing Lecture 5-6 Muhammad Ahmad Jan.
Jennifer Rexford Fall 2010 (TTh 1:30-2:50 in COS 302) COS 561: Advanced Computer Networks Energy.
© 2015 The SmartenIT Consortium 1 Commercial in Confidence DTM for minimization of inter-domain traffic cost Grzegorz Rzym, AGH, June 16, 2015 Socially-aware.
Organizations Are Embracing New Opportunities
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING CLOUD COMPUTING
C Loomis (CNRS/LAL) and V. Floros (GRNET)
6WIND MWC IPsec Demo Scalable Virtual IPsec Aggregation with DPDK for Road Warriors and Branch Offices Changed original subtitle. Original subtitle:
DI4R Conference, September, 28-30, 2016, Krakow
CSE 591: Energy-Efficient Computing Lecture 21 review
Cloud Computing Dr. Sharad Saxena.
Effective VM Sizing in Virtualized Data Centers
STEP VIRTUAL MACHINE MIGRATION FOR DYNAMIC RESOURCE ALLOCATION IN CLOUD COMPUTING ENVIRONMENT Guided By 2 2 STEP ParticipantsName Register Number K. Dileswara.
Presentation transcript:

© 2013 The SmartenIT Consortium 1 Commercial in Confidence Game Theoretic approach to energy efficiency Mateusz Wielgosz, Krzysztof Wajda, AGH Krakow Meeting, June 3, 2013 Socially-aware Management of New Overlay Application Traffic combined with Energy Efficiency in the Internet European Seventh Framework STREP FP ICT

© 2013 The SmartenIT Consortium 2 Commercial in Confidence Goal  Reduction of energy consumption via load aggregation and turning off physical machines.  Savings are considerable as idle server (in stand-by mode) can use up to 66% energy of fully-loaded machine.  Main stakeholders here are the IaaS providers, potential interference for users should be kept to a minimum.

© 2013 The SmartenIT Consortium 3 Commercial in Confidence Assumptions  We consider only single resource type on Physical Machines (PM) and all Virtual Machines (VM) seek this resource (e.g. processing power).  VMs are independent.  Cloud has sufficient capacity to satisfy all demands.  VM is to be hosted on single PM (VM/demand is not divisible)  Proposed solution is decentralized, but we show how it could be integrated with S-Box

© 2013 The SmartenIT Consortium 4 Commercial in Confidence Model and game  P = {P 1, P 2,…, P j,…, P |P| }, P j can supply C j resources.  V = {V 1, V 2,…, V i,…, V |V| }, V i demands D i resources.  Lj – total resource utilization of P j.  PM energy consumptionW j = ΣW i D i + W idle  VM „energy efficiency”A(P j )=D i /L j  Players: VMStrategy: pick PMUtility: „Energy efficiency”

© 2013 The SmartenIT Consortium 5 Commercial in Confidence Procedure  Physical Machines (PM):  Track your L j (sum current D i ’s)  Prompt housed VMs to check random PM whenever VM leav6es or joins this PM (alt. only when leaves)  (alt) Send L j to S-Box  Virtual Machines (VM):  Initial pick (random by default, alt. recommended by S-Box)  Check random PM when prompted by PM. (alt. check S-Box)  Migrate to another PM if energy efficiency there is lower by given threshold.  Threshold is used not only to aviod unnecessary migrations, but also to balance power savings and delay caused by migration.

© 2013 The SmartenIT Consortium 6 Commercial in Confidence Chain migration  Migration decreases energy efficiency cost at a local PM and increases efficiency at the remote PM.  D i /L loc D i /(L rem + D x )  Therefore migrating VM can leave new PM address behind, for next VMs. Address is wiped when first VM uses it, other VMs check random PM.  If remote PM still has enough resources it’s address will be left behind again by second VM.  If remote PM is „full”, then it is possible that other probing VMs will find PM with lower power cost.

© 2013 The SmartenIT Consortium 7 Commercial in Confidence Pros and Cons for S-Box involvement  Pros  Better performance  Better allocation from start of the procedure  Cons  Decentralized  No additional infrastructure  No synchronisation

© 2013 The SmartenIT Consortium 8 Commercial in Confidence To be adressed  Dynamics – potential area for social awareness.  Migration cost – power savings vs. user delay.

© 2013 The SmartenIT Consortium 9 Commercial in Confidence References  H. Khani, N. Yazdani, S. Mohammadi “Power-Aware Game for Cloud Computing” 6th International Symposium on Telecommunications,  D. Vesick, D. Tavangarian „Reducing Energy Consumption by Load Aggregation with an Optimized Dynamic Live Migration of Virtual Machines” 2010 International Conference on P2P, Parallel, Grid, Cloud and Internet Computing,  A. Gandhi, R. Das, M. Harchol-Balter, C. Lefurgy „Optimal Power Allocation in Server Farms” SIGMETRICS/Performance’09,  M. Harchol-Balter „Power Management in Data Centers” PROBE 2011.

© 2013 The SmartenIT Consortium 10 Commercial in Confidence Thank you ! Questions?

© 2013 The SmartenIT Consortium 11 Commercial in Confidence Backup slides 2 VM1 \ VM2Random PM betterRandom PM not better Remote PM not saturated(better, better)(better, try remote PM) Remote PM saturated(try random PM, better)(no improvement, no improvement) VM1 \ VM2Remote PM not saturatedRemote PM saturated Remote PM not saturated(better, better) Remote PM saturated(no improvement, better)(no improvement, no improvement)

© 2013 The SmartenIT Consortium 12 Commercial in Confidence Backup slides 1