Balanceamento de Carga em Clusters por Replicação de Dados Sob Demanda em um SGBDD
Resumo
In the last years, a significant growth of network traffic and processing load has been verified, mainly in database servers. Many factors have been responsible for this growth, such as: growth on demand for information availability, and increase of both size and quantity of stored information, as can be verified in multimedia application. One solution for the problem of processing large volume of information is the utilization of cluster based servers, working together with distributed database systems. This type of solution allows the parallel manipulation of great quantities of information. Although multiprocessing in clusters systems is, in fact, being used more and more due to high performance, low cost and high scalability characteristics, we verified that specific database systems for these architectures present some restrictions in relation to data replication, and load balance. The main objective of this work is the study and implementation of a mechanism that allows a better cluster load balance, in periods of high demand for information, by replicating the most requested data in a Distributed Database Management System (DDBMS). The complete process of copies generation is autonomous, with dynamic distribution of copies for the idle or less loaded node.
Referências
Jim Gray, Pat Helland, Patrick O'Neil, Dennis Shasha; The Dangers of Replication and a Solution. ACM-SIGMOD 1996, Conferencia Internacional de Gerenciamento de Dados, Quebec, 1996 pag. 173 - 182.
Senior Thesis by James S. Gwertzman; Autonomous Replication In Wide-Area Internetworks, Computer Science of Harward College, Cambridge Massachusets, 1995.
Ouri Wolfson, Sushil Jajodia, Yixiu Huang; An Adaptive Data Replicationn Algorithm, University of Illinois and NASA/CESDIS Goddard Space Flight Center-1997.
Gregory F. Pfister; In Search of Clusters - Norris Parker Smith, HPCwire (second edition - 1998) capítulos 1 , 4 e 10.
Chengzhong Xu - Nearest Neighbor Algorithms for load balancing in parallel computer - Dept. of Electrical & Computer Eng. of Wayne State University Detroit - 1996.
Azer Bestavros Speculative Data Dissemination and Service to reduce Server Load, Network Traffic and Service Time in Distributed Information Systems - Computer Science Department of Boston University - 1997.
Azer Bestavros - Load Profiling A Methodology for Scheduling Real-Time Tasks in a Distributed System - Computer Science Department of Boston University - 1996.
Master Thesis of Anastasia Anastasiadi - "A Study of Microeconomic Algorithms for Load Balancing and Data Replication in Distributed Computer Systems" - Dep. of Computer Science University of Grete - 1996.
ChengFu Chou, Leana Golubchik - A Performance Study of Dinamic Replication Techniques in Continuous Media Servers - Write Paper- Dept. of Computer Science University of Maryland at College Park - 1998.
Valeria Cardellini (Roma University), Michele Colajanni (Modena University), Philip S. Yu (Watson Research Center) - Dynamic Load Balancing on Web-Server Systema - Write Paper, IEEE Internet Computing magazine - may and june, 1999.
Sergio Takeo Kofuji, Martha Ximena Torres Delgado, Edward David Moreno Ordoñez, João Antônio Zuffo - Efeito da Migração de Páginas no SPADE - I, Um Multiprocessador de Larga Escala com Memória Compartilhada - Laboratório de Sistemas Integráveis, Escola Politécnica da Universidade de São Paulo - 1996.
