Improving the performance of a database system is one of the key research issues now a day. Distributed processing is an effective way to improve reliability and performance of a database system. Distribution of data is a collection of fragmentation, allocation and replication processes. Previous research works provided fragmentation solution based on empirical data about the type and frequency of the queries submitted to a centralized system. These solutions are not suitable at the initial stage of a database design for a distributed system. In this paper we have presented a fragmentation technique that can be applied at the initial stage as well as in later stages of a distributed database system for partitioning the relations. Allocation of fragments is done simultaneously in our algorithm. Result shows that proposed technique can solve initial fragmentation problem of relational databases for distributed systems properly.

unidentified trademarks mentioned in the paper are properties of their respective owners. Keywords: Content Addressable Techniques, Content Addressable Storage, Network

Abstract — Many data-intensive websites use databases that grow much faster than the rate that users access the data. Such growing datasets lead to ever-increasing space and performance overheads for maintaining and accessing indexes. Furthermore, there is often considerable skew with popular users and recent data accessed much more frequently. These observations led us to design Shinobi, a system which uses horizontal partitioning as a mechanism for improving query performance to cluster the physical data, and increasing insert performance by only indexing data that is frequently accessed. We present database design algorithms that optimally partition tables, drop indexes from partitions that are infrequently queried, and maintain these partitions as workloads change. We show a 60 × performance improvement over traditionally indexed tables using a real-world query workload derived from a traffic monitoring application I.

CCF-0326453, IIS-0133686, and CCF-0205544, as well as through the NASA AISR Program. The views and conclusions contained in this document are those of the author and should not be interpreted as representing the official policies, either expressed or implied, of any sponsoring institution,

Operations are effective: An attempt to apply machine learning to database parallelization by

Partitioning data is an attractive way to increase storage server throughput for web-like workloads, but two key challenges arise: (1) web workloads often do not have one clear partitioning and (2) it is challenging for the web developer to determine how to efficiently execute queries over partitioned tables. These two challenges can lead to per-query overhead, rather than useful work, dominating total throughput. This paper presents Dixie, a SQL query planner, optimizer, and executor for databases horizontally partitioned over multiple servers. Dixie automates the exploitation of tables with multiple copies partitioned in different ways, in order to increase throughput by expanding the portion of queries that need not be sent to all servers. Central to Dixie’s design are a cost model and plan generator that are mindful of queries small enough that query overhead may dominate the cost. We evaluate Dixie on a database and query stream taken from Wikipedia, partitioned across ten MySQL servers. By adding one copy of a 13 MB table and using Dixie’s query optimizer, we achieve a throughput improvement of 3.2X over a single optimized partitioning of each table and 8.5X over the same data on a single server. On specific queries Dixie with table copies increases throughput linearly with the number of servers, while the best single-table-copy partitioning achieves little scaling. For a large class of joins, which traditional wisdom suggests requires tables partitioned on the join keys, Dixie can find higher-performance plans using other partitionings. 1