This week's book giveaways are in the Refactoring and Agile forums. We're giving away four copies each of Re-engineering Legacy Software and Docker in Action and have the authors on-line! See this thread and this one for details.
I am in process of designing a database to hold archival data. The data is read only. No updates need to be made at any time. Entire details about a particular records spans across 10 to 12 rows in the database, if i go for relational approach. I need to display data for that entire record as part of display mechanism. The frequency of data to be retrieved is very less. maybe once in 4 months or so.
First Approach: Since the data is read only I figured that the entire data for a record can also be placed in a big table. Ideally that is something we always avoid, but since it is just read and display, if indexing is done properly, it should be ok right?
Second approach: put the entire record data in a xml file and save in the database as a blob. Use tools such as Db2 XML Extender for retreiving the data.
which of the two approaches should i go for. Are there any disadvantages in using second approach in terms of performances? Are there any advantages of second approach in terms of Database space(blob is compressed right?) I am quite new at this and will appreciate help. Thanks, Chetak