While it is not always possible or practical to select a DBMS before beginning the data management project, whenever possible the database itself should be tailored to the data. Attempting to "force fit" technical data into an unsuitable data management system will result in poor performance and user dissatisfaction.

Data storage requirements. When developing a new database system, it is sometimes difficult to estimate the data storage requirements. Even in existing systems, estimating future storage requirements is not always easy. As discussed elsewhere in this chapter, storage space is a function of data type, anticipated volume, and future expansions and additions to the database. In almost every case, initial estimates will be significantly lower than actual requirements. Some guidelines to consider when planning storage requirements are as follows.

Static database storage requirements are the easiest to deal with. Examining the data types that will be stored in the system, estimating the size of sample data populations, and considering the impact of future editing will make a reasonably accurate estimate possible.

Dynamic databases (active drilling and logging operations, new field acquisitions, expanding internationally, etc.) will be the most difficult to plan. In these situations, it is better to make a reasonable estimate and then design and build a system that can be expanded easily and inexpensively.

Data type greatly influences storage requirements. Large 3-D seismic volumes, high-resolution graphics and scanned images, and similar data types have a significant impact on storage...