Data compression, a multi-valuable database feature.

Data compression is a feature offered by modern databases. However, many companies do not use this feature in their database systems. Either they lack the knowledge of its existence in the design phase or due to budget issues. Some DBMSs offer features for an additional license price while other DBMSs offer it free of charge, e.g. might be included in the general package (open source DBMSs).

However, why should you use compression? In my opinion, it is fairly simple. The major benefits of compression are associated with space savings on disks which can range from 40 to 85 % (or more) depending on your data in the tables. Note that BLOB data does not compress well, or at all, as it is in most cases already compressed, e.g. JPEG, etc.. One beneficial facet of compression is the transfer of data into buffer pools “as-is” which gives increased hit ratio on the buffer pools and thus better buffer pool usage as more data is put into it.

By compressing data, your SQL queries will benefit on a larger scale as more data will exist in the memory at any given point in time. You will see overall performance improvements in your SQL workload which provide a reduction in CPU, execution time and I/O.

Another beneficial aspect lies in the DBMS backup-processes backing up tablespaces "as-is". In doing so, backup sizes become smaller as tables are compressed and as a result awarding you with larger space savings in the backup systems as well as shorter backup and restore times.

Also, transferring backups to and from the database host through your network will take a shorter time as will data transfer between database clusters.

As discussed above, the benefits of compression can have a very good and positive development of your database cost, especially if you have very large databases without compression. If you have a DBMS that requires an additional license for compression, then I highly recommend an evaluation and comparison of the license price versus the cost-savings in disk space and CPU. If you have a DBMS that has the compression feature included in the license, or you are using open source DBMS with this feature, then I strongly urge you to use this effective feature in reducing database costs.