Every business has data: from customer records to financial information, large volumes of data can be processed by companies daily, moving across platforms, applications, and systems, both internal and external. The reality is that companies, be they commercial entities or not-for-profit organisations, rely on data to support managerial decision making.

Businesses that appreciate the power of information technology utilize intelligent databases to store data, eliminating laborious manual processing. But databases themselves require regular maintenance to ensure that mission-critical data remains consistent, correct and reliable. It is also important to optimise databases - a process used to speed up the performance of data stores in handling data requests while improving the overall robustness of the system.

For example, Internet banking gives thousands of individuals access to their accounts, balances, and statements. If the back-end database did not support simultaneous access and request processing for all of the bank’s customers, we would simply stop using Internet banking or would switch to a different bank altogether.

Understanding database performance requirements is a key part of designing professional software, making database optimisation a very important exercise. Essentially, database optimisation revolves around these concepts:

• Scalability: The database should be designed to support increasing volumes of data and users as organisations grow in size

• Data processing speed: The database should process data with an optimal speed given the circumstances. The faster the system is able to manipulate data, the better.

• Concurrency: The database should be able to process multiple transactions issued by different users (Wikipedia: Concurrency Control)

• Normalization: The database should be logically structured and must not allow for data anomalies, for example storage of duplicate information. Normalization is used to preserve data integrity or completeness of data (Wikipedia: Data Integrity)

• Architecture: Depending on the purpose of the database, it should be optimised with an appropriate architecture in mind. For example, data warehouses require a column-oriented architecture, whereas Online Transaction Processing Systems are supported by row-oriented architectures.

The benefits of database optimisation are multiple, but can be summarized into two three broad categories:

• Business perspective: faster and better performing databases allow your business to get the information it needs quickly, saving you time and improving the quality of your decision making.

• Customer perspective: customers and trading partners expect prompt responses to their enquiries and requests. Optimised databases allow you to improve customer satisfaction through more informative, timely responses.

• IT perspective: well-managed systems are less likely to cause errors that could result in lost productivity or disruptions in the normal course of business.

You should consider database optimisation, if:

• Retrieving information from your database takes a lot of time, especially when multiple users are querying the system

• Your database has exhausted its resources and can no longer support new data structures or record new information about your business

• Your database has a lot of data that is repeated but should appear only once within the system

• Your database relies on legacy technology that slows downs your organization’s progress

• You want to share your database with your employees, customers and trading partners

• You database provides mission-critical information or must have robust availability with minimum downtime