As the database
evolves, youcan compare your initial estimates with emerging information to react
and plan more effectively. For example, suppose you estimate that a certain table will
be 5 GB in size initially and will grow at 3 GB per year, but when you are up and running
you discover that the table is actually 3 GB, and six months into production you
discover that it has grown to 8 GB. Youcan now revise your plans to reflect the
higher growth rate and thereby avoid space problems. Comparing production measures
of database size, growth, and usage patterns with your initial estimates will
provide valuable insights to help you avoid problems as you move forward. In this
way, documented guesses at an early stage are useful later on.
The same is true for key requirements such as availability and performance. If the
exact requirements are not clear, make some assumptions and document them.
These core requirements will heavily influence the decisions you make regarding
redundancy and capacity. As the system evolves and these requirements become
clearer, the history of these key decision criteria will be crucial in understanding the
choices that you made and will make in the future.
Pages:
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157