|
|
This is only a preview of the paper
Click here to register and get the full text.
Existing members click here to login
|
|
|
... This allows users of the database to achieve optimal performance from using indexing. ...
An Index according to OReilly is a conceptual part of a database table that may be used to speed up the retrieval of data from that table. ... To use B Trees tables should be reasonably large as well as badly balanced trees may inhibit performance. ...
With testing the performance of indexing Vs full table scan there only needs to be one table made available for testing and seven attributes that are included in the table will be sufficient enough to carry out this study. ...
Test Strategy
Goals
The goal of this study is to compare the different performance speeds of different indexing methods to the performance of a full table scan. ... The service that will be offered by the system will be the ability of the index to optimise the best performance of the data that is required. ...
Metrics
The performance of these indexing methods will depend on the amount of data that is being used and extracted and the size of the table. ... Other metrics include:
The amount of table space allowed by the network
The amount of index space allowed by the network
The amount of time to execute five consecutive queries on each indexing method
The amount of bytes needed to be sent across the network as this might slow down the query
Parameters
The parameters that will affect the performance of the queries are as follows:
The amount of users already on the network
The time of the day
The attributes that have been specified in the query
The type of index used
The amount of data included in the table
The amount of data output from the query
The amount of time it takes to create the index
The size of each column in the database
Factors
The data for this experiment will contain one large table with 7 attributes. ... The queries that will be used will be run 5 times to get an accurate time of the performance of the index. ...
Comparison of Statistics
When performing a full table scan against an index at 2% and 5% there are a lot more physical reads from the full table scan therefore slowing the query performance. ...
Overall the experiment that was carried out supported most things that were discussed in the hypothesis full table scans are proved to perform faster when accessing lots of rows in the table and an index will perform better when accessing lesser rows in the database.
Approximate Word count = 4549
Approximate Pages = 18.2
(250 words per page double spaced)
|
|
|
|
|
|