A performance comparison of normal query and adaptive query in Web-client monitoring system

摘要

The Internet access restriction is difficult to be handled, because it depends on user behavior or mentality. Limitation of Internet access does not have significant impact for long time, because the increase of limitation user access increase attempts to crack the access restriction. We assumed that "morality filter" should be more effective rather than existing filter. In this research the monitoring system was design for presenting the traffic-distribution data from user Internet environment. Adaptive-query searching algorithm is implemented using Eddy scheme that consists of Pipeline-Join and Hash-Join operation for searching the data based on criterion. This criterion is sent by the user using monitoring homepage on Web server. From this experiment, the aggregation-time of adaptive-query have a better performance for amount data more than 1000 tuples, but for data of up to 1000 using normal-query is better.