Block Based Resumption Techniques for Efficient Handling of Unsuccessful Loads in Data Warehouse

Block Based Resumption Techniques for Efficient Handling of Unsuccessful Loads in Data Warehouse

연구 분야: Databases

학회: International Conference on Advances in Computing and Data Sciences

ETL is the acronym for extract transform and load, it’s a process to extract, transform and load data into the data warehouse from sources that could be other transactional database, text files, logs etc. This source could be heterogeneous or homogeneous. Any of the steps in ETL could fail and measures are required to resume the process. There could be several reasons for failure like network break down, hard ware crash, database unavailable, over loaded systems, etc. The issue of ETL failure is a serious one because it’s a time consuming processes. In case of failure of ETL what should be the strategy to resume the process so that the focus of resumption is on the data that failed to load rather than on the data that is already in the data warehouse. In this paper a block based approach is used to load the data that failed during load of ETL. Empirical results show the block based approach performs better in terms of resumption time as compared to SQL EXCEPT.