Research On Join Query Processing Schemes In A Mobile Computing Environment

A mobile computing environment is an inevitable trend of the network, and applications under the environment are becoming increasingly common. Query processing is one of the most basic technologies for mobile computing, while join processing, the most commonly-used operation in query, causes the highest time-and-space consuming, also greatly makes the internet communication cost's burst. Consequently, against the mobile environment which is highly restricted by battery energy, communication bandwidth, memory capacity and handling ability etc, to select one join processing scheme properly is extremely important.Based on the discussion of asymmetric features of mobile computing environment and the result relation's estimate for join operation in relational database, the cost estimate methods for join query processing schemes in the environment are presented. The costs of five join processing schemes are evaluated from data transmission and energy consumption, and the performances of them are comparatively analyzed. Then seven realistic qualities are achieved to guide us to select the suitable processing scheme, so as to reduce the amount of data transmission and energy consumption at mobile units, moreover to drop the cost of communication. The research is carried out by combing theory analysis and simulation experiments. The wireless local area network is build with the help of a wireless access point. The experiments fully demonstrate that the estimate methods and the qualities are reliable, and more applicable in comparison with the existing similar models and results.The major achievements in this paper are listed here: 1. A new asymmetric feature of energy consumption at a mobile unit when inserting the data received in a database or not is exploited on the wide investigation of network structure and features and so on under the mobile computing environment. It is very important for the estimate of energy consumption at mobile units.2. The estimate thought of result size for query operations in relational database theory is introduced into the operation cost estimate for join query processing schemes under the mobile computing environment.3. The costs of three single-join processing schemes are evaluated from data transmission and energy consumption on the basis of the former cost estimate method, and the performances of them are comparatively analyzed. All the results are achieved to guide us to select the suitable single-join query processing scheme.4. The costs of two multi-join processing schemes are evaluated on the basis of the former cost estimate method, and the performances of them are comparatively analyzed so as to guide us to select easily the suitable multi-join query processing scheme.5. A great deal of simulation experiments are carried out under the wireless local area network. They verify the validity of the operation cost estimate methods and the performance compare for join query processing schemes. At the same time, the influences of some selectivity factors to the performance of single-join query processing schemes are analyzed on the basis of the test results, as well as the efficiency of letting the server choose freely the processing schemes for multi-join query.