Cost Tradeoff of Consistency Over Data Centers Ozlem Bilgir.

Slides:



Advertisements
Similar presentations
Utility Optimization for Event-Driven Distributed Infrastructures Cristian Lumezanu University of Maryland, College Park Sumeer BholaMark Astley IBM T.J.
Advertisements

Ronald J. Zimmer, CAE President & CEO Continental Automated Buildings Association (CABA) HOME AREA NETWORKS in a Smart Grid.
Hadi Goudarzi and Massoud Pedram
Neural and Evolutionary Computing - Lecture 4 1 Random Search Algorithms. Simulated Annealing Motivation Simple Random Search Algorithms Simulated Annealing.
Walter Binder University of Lugano, Switzerland Niranjan Suri IHMC, Florida, USA Green Computing: Energy Consumption Optimized Service Hosting.
4 good reasons why Energy Efficiency is Important.
A Cyber-Physical Systems Approach to Energy Management in Data Centers Presented by Chen He Adopted form the paper authors.
Green Cloud Computing Hadi Salimi Distributed Systems Lab, School of Computer Engineering, Iran University of Science and Technology,
GREEN CLOUD By Sphoorthy. LOGO WHAT IS CLOUD COMPUTING? Cloud computing is a model for enabling convenient, on- demand network access to a shared pool.
RELATED BACKGROUND WORK OZLEM BILGIR. OUTLINE 1- Gandhi et al., Optimal Power Allocation in Server Farms, SIGMETRICS’09 2-Chen et al., Managing Server.
GREEN DATA CENTERS : LOAD BALANCING AND ENERGY MANAGEMENT OZLEM BILGIR.
Control System for Energy Efficient Data Centers Ozlem Bilgir.
Exploiting Content Localities for Efficient Search in P2P Systems Lei Guo 1 Song Jiang 2 Li Xiao 3 and Xiaodong Zhang 1 1 College of William and Mary,
1 CAPS: A Peer Data Sharing System for Load Mitigation in Cellular Data Networks Young-Bae Ko, Kang-Won Lee, Thyaga Nandagopal Presentation by Tony Sung,
Transportation Models Transportation problem is about distribution of goods and services from several supply locations to several demand locations. Transportation.
Zoë Abrams, Ashish Goel, Serge Plotkin Stanford University Set K-Cover Algorithms for Energy Efficient Monitoring in Wireless Sensor Networks.
Intelligent Placement of Datacenters for Internet Services Íñigo Goiri, Kien Le, Jordi Guitart, Jordi Torres, and Ricardo Bianchini 1.
1 Algorithms for Bandwidth Efficient Multicast Routing in Multi-channel Multi-radio Wireless Mesh Networks Hoang Lan Nguyen and Uyen Trang Nguyen Presenter:
Grant Klinger Abdulrahman Maddi Bill Miller Taylor Miller.
Common Energy Mistakes
Client-Server Assignment for Internet Distributed Systems.
6.8 –Systems of Inequalities. Just like systems of equations, but do the inequality part!
Building Sustainable MIS Infrastuctures
Advanced Energy Management in Cloud Computing multi data center environments Giuliana Carello, DEI, Politecnico di Milano Danilo.
Cost- and Energy-Aware Load Distribution Across Data Centers Presented by Shameem Ahmed Kien Le, Ricardo Bianchini, Margaret Martonosi, and Thu D. Nguyen.
What is a Network?. Definition of a computer network A computer network is a system in which computers are connected to share information and resources.
E-STAB: Energy-Efficient Scheduling for Cloud Computing Applications with Traffic Load Balancing Dzmitry KliazovichUniversity of Luxembourg, Luxembourg.
OPTIMAL SERVER PROVISIONING AND FREQUENCY ADJUSTMENT IN SERVER CLUSTERS Presented by: Xinying Zheng 09/13/ XINYING ZHENG, YU CAI MICHIGAN TECHNOLOGICAL.
Cloud Computing Energy efficient cloud computing Keke Chen.
胡秩瑋.  INTRODUCTION  RELATED WORK  FORMULATION AND MODELING  SOLUTION METHOD DESIGN  ELECTRICITY PRICE AT CERTAIN LOCATIONS FOR GOOGLE INTERNET DATA.
Building Green Cloud Services at Low Cost Josep Ll. Berral, Íñigo Goiri, Thu D. Nguyen, Ricard Gavaldà, Jordi Torres, Ricardo Bianchini.
EPA Activities - Tires Lois Platte September 19, 2002.
Michigan’s Energy Future Today Robert Jackson DELEG.
LIEN – Reporting Energy & CO 2 Emissions for Carbon Tax and Emissions Trading. Fuel Use and Distribution in Ireland.
Optimal Client-Server Assignment for Internet Distributed Systems.
The Smart Grid: A Brief Introduction Qinran Hu Ph.D. Candidate Jun 12 th, 2014 Knoxville, Tennessee.
1 Optimal Resource Placement in Structured Peer-to-Peer Networks Authors: W. Rao, L. Chen, A.W.-C. Fu, G. Wang Source: IEEE Transactions on Parallel and.
Summer Report Xi He Golisano College of Computing and Information Sciences Rochester Institute of Technology Rochester, NY
1 Distributed Energy-Efficient Scheduling for Data-Intensive Applications with Deadline Constraints on Data Grids Cong Liu and Xiao Qin Auburn University.
5.4 – Solving Compound Inequalities. Ex. Solve and graph the solution.
Shipping to Streaming: Is this shift green? Kevin Leeds.
Joint Power Optimization Through VM Placement and Flow Scheduling in Data Centers DAWEI LI, JIE WU (TEMPLE UNIVERISTY) ZHIYONG LIU, AND FA ZHANG (CHINESE.
PAPER PRESENTATION Real-Time Coordination of Plug-In Electric Vehicle Charging in Smart Grids to Minimize Power Losses and Improve Voltage Profile IEEE.
Downlink Scheduling With Economic Considerations to Future Wireless Networks Bader Al-Manthari, Nidal Nasser, and Hossam Hassanein IEEE Transactions on.
Mobile Agent Migration Problem Yingyue Xu. Energy efficiency requirement of sensor networks Mobile agent computing paradigm Data fusion, distributed processing.
Managing Server Energy and Operational Costs Chen, Das, Qin, Sivasubramaniam, Wang, Gautam (Penn State) Sigmetrics 2005.
11.1 Ratios and Proportions Solve proportions. Proportion – equates two ratios extreme mean This proportion is read as “a is to b as c is to d.” You must.
Database replication policies for dynamic content applications Gokul Soundararajan, Cristiana Amza, Ashvin Goel University of Toronto Presented by Ahmed.
Efficient Content Sharing Taking Account of Updating Replicas in Hybrid Peer-to-Peer Networks Tatsuru Kato, Shinji Sugawara, Yutaka Ishibashi Nagoya Institute.
Energy Efficiency Potential in the Wisconsin Industrial Sector A Discussion With the Wisconsin Industrial Energy Group November 6, 2008.
Online Assignment under SLAs and Brown Energy Caps Spyridon Antonakopoulos 2 nd NSF Workshop on the Science of Power Management, August 18 th 2010 Inspired.
Feifei Chen Swinburne University of Technology Melbourne, Australia
Data Consolidation: A Task Scheduling and Data Migration Technique for Grid Networks Author: P. Kokkinos, K. Christodoulopoulos, A. Kretsis, and E. Varvarigos.
Rough-Cut Capacity Planning in SCM EIN 5346 Logistics Engineering Fall, 2015.
An Energy-Efficient Approach for Real-Time Tracking of Moving Objects in Multi-Level Sensor Networks Vincent S. Tseng, Eric H. C. Lu, & Kawuu W. Lin Institute.
CSE 591: Energy-Efficient Computing Lecture 3 SPEED: processor Anshul Gandhi 347, CS building
On the Placement of Web Server Replicas Yu Cai. Paper On the Placement of Web Server Replicas Lili Qiu, Venkata N. Padmanabhan, Geoffrey M. Voelker Infocom.
Implementation of Simple Cloud-based Distributed File System Group ID: 4 Baolin Wu, Liushan Yang, Pengyu Ji.
1 Terrain-Constrained Mobile Sensor Networks Shu Zhou 1, Wei Shu 1, Min-You Wu 2 1.The University of New Mexico 2.Shanghai Jiao Tong University IEEE Globecom.
Rough-Cut Capacity Planning in SCM Theories & Concepts
1 Data Center Sizes and Classes: Overview Scott Schuetz, Associate Director, IT May 7, 2013.
Online Data Storage Companies MY Docs Online. Comparison Name Personal Edition Enterprise Edition Transcription Edition Price $9.95 monthly rate $4.99.
Do you know? By 2015, an estimated 40-48% of new non-residential construction by value will be green, equating to a $ billion.
Alternative Search Formulations and Applications
Adel Nadjaran Toosi and Rajkumar Buyya
Leontief Input-Output Model
Introduction to CAST Technical Support
Servers Options Put all services on one server, or
Solution methods for NP-hard Discrete Optimization Problems
Lecture 9: Allocation of costs to products
Presentation transcript:

Cost Tradeoff of Consistency Over Data Centers Ozlem Bilgir

Outline Problem What do we need to solve the problem? What is Consistency? General Formulation Solution Approaches Results

Source: EPA Report to Congress on Server and Data Center Energy Efficiency What is the Problem? In 2006, data centers in US consumed ~60billion KWhs I.Cost : In 2006, total data center electricity cost in US was $4.5billion II.Greenhouse Gases : In 2008, data centers caused more carbon emissions than Argentina &Netherland (Source: McKinsey & Company, 2008)

What Do We Need? Good way of request distribution : to minimize cost Brown Energy Caps: to limit the brown energy consumption at data centers Satisfy SLA Don’t overload any of the data center Satisfy consistency over replicas

What is Consistency? Data are not stored only in one data center 3 types of consistency; – Weak Consistency: ex. VOiP – Eventual Consistency: ex. – Strong Consistency: ex. File upload

General Formulation

Assumptions 3 data center -- They are all replicas of each other All data centers can serve all types of requests Strong consistency over data centers Data centers consume same energy to process all types of requests

Solution Approaches Simulated Annealing: To solve optimization problem CA-Heuristic : – Sort the data centers according to Cost i (t)/CDF i (L;LCi) ratio, – send to first one until it overloads, – send to next ones until SLA satisfies, – send to cheapest one CU-Heuristic : Don’t take cost into account

Results

Conclusion Extra cost from consistency is added to the general cost Future Work: Back to the Control Theory!!!