Design and Performance Evaluation of Queue-and-Rate-Adjustment Dynamic Load Balancing Policies for Distributed Networks Zeng Zeng, Bharadwaj, IEEE TRASACTION.

Slides:



Advertisements
Similar presentations
Ch. 12 Routing in Switched Networks
Advertisements

Dr. Kalpakis CMSC 621, Advanced Operating Systems. Distributed Scheduling.
Scheduling in Web Server Clusters CS 260 LECTURE 3 From: IBM Technical Report.
Ch. 12 Routing in Switched Networks Routing in Packet Switched Networks Routing Algorithm Requirements –Correctness –Simplicity –Robustness--the.
Resource Management §A resource can be a logical, such as a shared file, or physical, such as a CPU (a node of the distributed system). One of the functions.
The Selective Intermediate Nodes Scheme for Ad Hoc On-Demand Routing Protocols Yunjung Yi, Mario gerla and Taek Jin Kwon ICC 2002.
Min Song 1, Yanxiao Zhao 1, Jun Wang 1, E. K. Park 2 1 Old Dominion University, USA 2 University of Missouri at Kansas City, USA IEEE ICC 2009 A High Throughput.
Load Rebalancing for Distributed File Systems in Clouds Hung-Chang Hsiao, Member, IEEE Computer Society, Hsueh-Yi Chung, Haiying Shen, Member, IEEE, and.
Cooperative Overlay Networking for Streaming Media Content Feng Wang 1, Jiangchuan Liu 1, Kui Wu 2 1 School of Computing Science, Simon Fraser University.
DYNAMIC POWER ALLOCATION AND ROUTING FOR TIME-VARYING WIRELESS NETWORKS Michael J. Neely, Eytan Modiano and Charles E.Rohrs Presented by Ruogu Li Department.
The JOURNEY Active Network Model Maximilian Ott et al. IEEE Journal on Selected Areas in Communications, vol.19, no. 3, March 2001.
A Parallel Computational Model for Heterogeneous Clusters Jose Luis Bosque, Luis Pastor, IEEE TRASACTION ON PARALLEL AND DISTRIBUTED SYSTEM, VOL. 17, NO.
Effectively Utilizing Global Cluster Memory for Large Data-Intensive Parallel Programs John Oleszkiewicz, Li Xiao, Yunhao Liu IEEE TRASACTION ON PARALLEL.
1 Peer-To-Peer-Based Resource Discovery In Global Grids: A Tutorial Rajiv Ranjan, Aaron Harwood And Rajkumar Buyya, The University Of Melbbourne IEEE Communications.
Dynamic Load Balancing Experiments in a Grid Vrije Universiteit Amsterdam, The Netherlands CWI Amsterdam, The
A Trust Based Assess Control Framework for P2P File-Sharing System Speaker : Jia-Hui Huang Adviser : Kai-Wei Ke Date : 2004 / 3 / 15.
Fault-tolerant Adaptive Divisible Load Scheduling Xuan Lin, Sumanth J. V. Acknowledge: a few slides of DLT are from Thomas Robertazzi ’ s presentation.
Design, Implementation, and Evaluation of Differentiated Caching Services Ying Lu, Tarek F. Abdelzaher, Avneesh Saxena IEEE TRASACTION ON PARALLEL AND.
1 Introduction to Load Balancing: l Definition of Distributed systems. Collection of independent loosely coupled computing resources. l Load Balancing.
Grid Load Balancing Scheduling Algorithm Based on Statistics Thinking The 9th International Conference for Young Computer Scientists Bin Lu, Hongbin Zhang.
Strategies for Implementing Dynamic Load Sharing.
Maximizing the Lifetime of Wireless Sensor Networks through Optimal Single-Session Flow Routing Y.Thomas Hou, Yi Shi, Jianping Pan, Scott F.Midkiff Mobile.
12006/9/26 Load Balancing in Dynamic Structured P2P Systems Brighten Godfrey, Karthik Lakshminarayanan, Sonesh Surana, Richard Karp, Ion Stoica INFOCOM.
A Multi-Agent Learning Approach to Online Distributed Resource Allocation Chongjie Zhang Victor Lesser Prashant Shenoy Computer Science Department University.
On Fairness, Optimizing Replica Selection in Data Grids Husni Hamad E. AL-Mistarihi and Chan Huah Yong IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS,
1 Incentive-Based Scheduling for Market-Like Computational Grids Lijuan Xiao, Yanmin Zhu, Member, IEEE, Lionel M. Ni, Fellow, IEEE, and Zhiwei Xu, Senior.
A Load Balancing Framework for Adaptive and Asynchronous Applications Kevin Barker, Andrey Chernikov, Nikos Chrisochoides,Keshav Pingali ; IEEE TRANSACTIONS.
Dynamic Load Sharing and Balancing Sig Freund. Outline Introduction Distributed vs. Traditional scheduling Process Interaction models Distributed Systems.
Web Server Load Balancing/Scheduling Asima Silva Tim Sutherland.
Distributed Quality-of-Service Routing of Best Constrained Shortest Paths. Abdelhamid MELLOUK, Said HOCEINI, Farid BAGUENINE, Mustapha CHEURFA Computers.
Exploring VoD in P2P Swarming Systems By Siddhartha Annapureddy, Saikat Guha, Christos Gkantsidis, Dinan Gunawardena, Pablo Rodriguez Presented by Svetlana.
1 Distributed Operating Systems and Process Scheduling Brett O’Neill CSE 8343 – Group A6.
OPTIMAL SERVER PROVISIONING AND FREQUENCY ADJUSTMENT IN SERVER CLUSTERS Presented by: Xinying Zheng 09/13/ XINYING ZHENG, YU CAI MICHIGAN TECHNOLOGICAL.
 Escalonamento e Migração de Recursos e Balanceamento de carga Carlos Ferrão Lopes nº M6935 Bruno Simões nº M6082 Celina Alexandre nº M6807.
Improving Capacity and Flexibility of Wireless Mesh Networks by Interface Switching Yunxia Feng, Minglu Li and Min-You Wu Presented by: Yunxia Feng Dept.
A Prediction-based Fair Replication Algorithm in Structured P2P Systems Xianshu Zhu, Dafang Zhang, Wenjia Li, Kun Huang Presented by: Xianshu Zhu College.
1 Multiprocessor and Real-Time Scheduling Chapter 10 Real-Time scheduling will be covered in SYSC3303.
Maximum Network Lifetime in Wireless Sensor Networks with Adjustable Sensing Ranges Cardei, M.; Jie Wu; Mingming Lu; Pervaiz, M.O.; Wireless And Mobile.
Computer Networks with Internet Technology William Stallings
CSCI 465 D ata Communications and Networks Lecture 15 Martin van Bommel CSCI 465 Data Communications & Networks 1.
Parallelization of Classification Algorithms For Medical Imaging on a Cluster Computing System 指導教授 : 梁廷宇 老師 系所 : 碩光通一甲 姓名 : 吳秉謙 學號 :
TELE202 Lecture 6 Routing in WAN 1 Lecturer Dr Z. Huang Overview ¥Last Lecture »Packet switching in Wide Area Networks »Source: chapter 10 ¥This Lecture.
MMAC: A Mobility- Adaptive, Collision-Free MAC Protocol for Wireless Sensor Networks Muneeb Ali, Tashfeen Suleman, and Zartash Afzal Uzmi IEEE Performance,
CS 484 Load Balancing. Goal: All processors working all the time Efficiency of 1 Distribute the load (work) to meet the goal Two types of load balancing.
Design Issues of Prefetching Strategies for Heterogeneous Software DSM Author :Ssu-Hsuan Lu, Chien-Lung Chou, Kuang-Jui Wang, Hsiao-Hsi Wang, and Kuan-Ching.
DISTRIBUTED COMPUTING
Ching-Ju Lin Institute of Networking and Multimedia NTU
Energy-Efficient Randomized Switching for Maximizing Lifetime in Tree- Based Wireless Sensor Networks Sk Kajal Arefin Imon, Adnan Khan, Mario Di Francesco,
An Adaptive Zone-based Storage Architecture for Wireless Sensor Networks Thang Nam Le, Dong Xuan and *Wei Yu Department of Computer Science and Engineering,
Global Clock Synchronization in Sensor Networks Qun Li, Member, IEEE, and Daniela Rus, Member, IEEE IEEE Transactions on Computers 2006 Chien-Ku Lai.
Spring 2000CS 4611 Routing Outline Algorithms Scalability.
Ensieea Rizwani An energy-efficient management mechanism for large-scale server clusters By: Zhenghua Xue, Dong, Ma, Fan, Mei 1.
An adaptive Multihop Clustering Scheme for Highly Mobile Ad Hoc Networks Tomoyuki Ohta, Shinji Inoue, and Yoshiaki Kakuda ISADS 2003.
Critical Area Attention in Traffic Aware Dynamic Node Scheduling for Low Power Sensor Network Proceeding of the 2005 IEEE Wireless Communications and Networking.
Packet Classification Using Dynamically Generated Decision Trees
A Two-Tier Heterogeneous Mobile Ad Hoc Network Architecture and Its Load-Balance Routing Problem C.-F. Huang, H.-W. Lee, and Y.-C. Tseng Department of.
Grid Performability, Modelling and Measurement AHM’04 Optimal Tree Structures for Large-Scale Grids J. Palmer I. Mitrani School of Computing Science University.
Deploying Sensors for Maximum Coverage in Sensor Network Ruay-Shiung Chang Shuo-Hung Wang National Dong Hwa University IEEE International Wireless Communications.
Distributed Scheduling Motivations: reduce response time of program execution through load balancing Goal: enable transparent execution of programs on.
Courtesy Piggybacking: Supporting Differentiated Services in Multihop Mobile Ad Hoc Networks Wei LiuXiang Chen Yuguang Fang WING Dept. of ECE University.
HAT: Heterogeneous Adaptive Throttling for On-Chip Networks Kevin Kai-Wei Chang Rachata Ausavarungnirun Chris Fallin Onur Mutlu.
William Stallings Data and Computer Communications
Web Server Load Balancing/Scheduling
Web Server Load Balancing/Scheduling
Introduction to Load Balancing:
Dynamic Graph Partitioning Algorithm
Distributed Energy Efficient Clustering (DEEC) Routing Protocol
Hawk: Hybrid Datacenter Scheduling
Job-aware Scheduling in Eagle: Divide and Stick to Your Probes
Performance-Robust Parallel I/O
Presentation transcript:

Design and Performance Evaluation of Queue-and-Rate-Adjustment Dynamic Load Balancing Policies for Distributed Networks Zeng Zeng, Bharadwaj, IEEE TRASACTION ON COMPUTERS, VOL. 55, NO. 11, NOVEMBER 2006 Presented by 張肇烜

Outline Introduction Classification of Dynamic Load Balancing Algorithms Comparative Study on the Algorithms Performance Evaluation and Discussions Extension to Large Scale Cluster Systems Conclusions

Introduction Centralized dynamic load balancing. Scheduler can handle most of the communication and computation overheads efficiently.

Introduction (cont.) Distributed dynamic load balancing. More advantages, such as scalability, flexibility, and reliability.

Introduction (cont.) A distributed computer system.

Introduction (cont.) In this paper, we classify the dynamic distributed load balancing algorithms for heterogenous distributed computer systems into three policies: Queue Adjustment Policy (QAP) Rate Adjustment Policy (RAP) Queue and Rate Adjustment Policy (QRAP)

Introduction (cont.) QAP: Estimated Load Information Scheduling Algorithm (ELISA). Perfect Information Algorithm (PIA). RAP: Rate-based Load Balancing via Virtual Routing (RLBVR).

Introduction (cont.) QRAP: Queue-based Load Balancing via Virtual Routing (QLBVR).

Classification of Dynamic Load Balancing Algorithms Queue Adjustment Policy:

Classification of Dynamic Load Balancing Algorithms (cont.) Rate Adjustment Policy:

Classification of Dynamic Load Balancing Algorithms (cont.) Queue and Rate Adjustment Policy:

Comparative Study on the Algorithms In distributed dynamic load balancing algorithms, the nodes in the system exchange their status information at a periodic interval of time T s,which is called the status exchange interval. The instant at which this information exchange takes place is called a status exchange epoch.

Comparative Study on the Algorithms (cont.) Each status exchange interval is further divided into equal subintervals denoted as estimation intervals, T e. The points of division are called estimation epochs.

Comparative Study on the Algorithms (cont.) Intervals of estimation and status exchange.

Comparative Study on the Algorithms (cont.) ELISA: Each node computes the average load on itself and its neighboring nodes. Nodes in the neighboring set whose estimated queue length is less than the estimated average queue length by more than a threshold θ form an active set.

Comparative Study on the Algorithms (cont.) ELISA: The node under consideration transfers jobs to the nodes in the active set until its queue length is not greater than θ and more than the estimated average queue length.

Comparative Study on the Algorithms (cont.) RLBVR:

Comparative Study on the Algorithms (cont.) QLBVR caries out coarse adjustment on job transferring and processing rates and fine adjustment on queue length. Coarse adjustment (on transfer and processing rates). Fine adjustment (on queue lengths).

Comparative Study on the Algorithms (cont.) QLBVR: When the job incoming rates change slightly, coarse adjustment can work well. When the system load is very high and job incoming rates change rapidly, fine adjustment can balance the queue lengths in a short time.

Performance Evaluation and Discussions Effect of system loading:

Performance Evaluation and Discussions (cont.) When the load of the system is light or moderate, RLBVR and QLBVR have a better performance than ELISA. When the rate of jobs becomes high, ELISA and QLBVR have a much better performance than RLBVR.

Performance Evaluation and Discussions (cont.) Effect of T s :System loading is light.

Performance Evaluation and Discussions (cont.) Effect of T s :System loading is moderate.

Performance Evaluation and Discussions (cont.) Effect of T s :System loading is moderate.

Extension to Large Scale Cluster Systems The mesh-connected cluster system.

Extension to Large Scale Cluster Systems (cont.) Mean response time of jobs for five different algorithms under different system utilization. System utilization is light or moderate. System utilization is high.

Extension to Large Scale Cluster Systems (cont.) System utilization is light or moderate.

Extension to Large Scale Cluster Systems (cont.) System utilization is high.

Extension to Large Scale Cluster Systems (cont.) Experiments when the arrival of loads is varying rapidly.

Extension to Large Scale Cluster Systems (cont.)

Conclusion With our rigorous experiments, we have shown that, when the system loads are light or moderate, algorithms of the Rap policy are preferable with longer T s. When the system loads are fairly high, QAP policy and QRAP policy have better performance than RAP policy.