Of those zillion of bytes of data it would be interesting to determine what amount of that data is in relational databases incased in products like SQL Server. I would venture that the percentage of data in RDBMs format is not what most of us might think. There is a new technology being used that I believe is called noSQL being a data storage system that does not use an RDBMs.

With the need for larger more searchable data stores that are not the traditional row, field, table approach to a data model, new technologies are coming to the forefront. These data stores do now hold and will continue to hold in the future more data then we see in the traditional RDBMs. Since this technology is emerging and they are just starting to understand searching of the vast numbers of related clusters within a data store there is little chance that those stores will be encrypted over the short term.

That leaves us with the question of compression of data within an RDBMs only. And the key issue as I see it centers around if the speed of searching + the speed of retrieval + the speed of compression and or decompression is acceptable to the user. And if the complexity of performing the required activities by the programmer and DBA are not overly difficult. Others can answer that.

Should we compress, possibly or the classic answer of "It depends" might be appropriate.

M.