Data administration is that function concerned with the management, planning and documentation of the data resource of some organization.
Data administration is concerned with the management of an organization’s meta-data, that is, data about data. It is a function which deals with the conceptual or business view of an organization’s data resource.
The key concept in the move towards data administration has been that data, like capital, personnel etc., should be treated as a manageable resource. In other words, data is a critical commodity in an organization’s attempt to compete in the open marketplace. In this sense, the data administration function is seen as a key part of the data and information management strategy of organizations. The scope of the data administration function will vary according to the organization.
There are however a core set of activities for which a data administration function is typically responsible:
Consultancy – offering consultancy on all aspects related to an organization’s meta-data, particularly expertise in data analysis.
i. Corporate awareness – educating to increase awareness of the importance of data; also disseminating information on what data exists and for what purpose
ii. Corporate requirements – identifying corporate data requirements, particularly building a corporate data architecture which incorporates strategic planning
iii. Data analysis – coordinating the use of a standard data analysis methodology, and using such a methodology to develop business data models
iv. Data control – implementing standards for ensuring that access to data is controlled; also ensuring that suitable recovery procedures are in place
v. Data security – ensuring that both technical and administrative controls are in place to protect against threats
vi. Data definition – implementing standards for the definition of data and controlling the medium for the recording and communication of such definitions
vii. Data integrity – implementing standard mechanisms for ensuring the integrity of an organization’s data; also documenting the rules for ensuring integrity
viii. Data dictionary management – promoting the use of a logical data dictionary and implementing standards for its control; also monitoring the use of and content of the data dictionary
ix. Data privacy – implementing procedures to ensure that the organization complies with any legislation concerning national data regulation
x. Data sharing – encouraging sharing of data across applications and promoting the idea of data that is independent of applications
7.2 The Need for Data Administration
The need for a data administration function arises from the problems of managing data within an organization. Such problems include:
i. A number of applications are developing within some organizations which use different definitions for the same data
ii. Data held by a number of diverse applications is inconsistent
iii. Decision-makers within an organization receive conflicting data from different sources within that organization
iv. Decision-makers receive data too late for it to be useful
v. Decision-makers receive too much irrelevant data
vi. There are notable gaps in the data collected by an organization
vii. Departments within an organization have no clear idea why they collect certain data
In this view, data administration may be seen as an attempt to develop some order from the chaos of corporate information systems. However, data administration also involves planning the data required for future information systems.
The Difficulties of Data Administration
It is rare for a data administration function to be set up in an environment in which no applications systems yet exist. Most data administration departments are set up as a response to some of the problems outlined above. In such a situation it is usually a significant task to reconcile data collected from diverse areas of the organisation. Many parts of the organisation will openly disagree about data requirements. Data administration is hence frequently a function which is shaped by the sociology, politics and economics of organisations.
7.4 Database Administrator
The database administrator (DBA) is responsible for the technical implementation of database systems, managing the database systems currently in use, and setting and enforcing policies for their use. Whereas the data administrator works primarily at the conceptual level of business data, the DBA works primarily at the physical level. The place where the data administrator and the DBA meet is at the logical level. Both the data administrator and DBA must be involved in the system-independent specification and management of data.
The need for a specialist DBA function varies depending on the size of the database system being administered. In terms of a small desktop database system, the main user of the database will probably perform all DBA tasks such as taking regular backups of data. However, when a database is being used by many people and the volume of data is significant, the need for a person or persons who specialize in administration functions becomes apparent.
Key Functions of Database Administration
In many organizations, particularly small organizations, the DBA will be expected to undertake many data administration tasks. However, in general, the DBA can be said to have the following core responsibilities: administration of the database, administration of the DBMS and administration of the database environment.
Administration of the Database
The DBA would normally be expected to engage in the following key activities in relation to administering a database or a series of databases:
i. Physical design – whereas the data administrator will be concerned with the conceptual and logical design of database systems, the DBA will be concerned with the physical design and implementation of databases
ii. Data standards and documentation – ensuring that physical data is documented in a standard way such that multiple applications and end-users can access the data effectively
iii. Monitoring data usage and tuning database structures – monitoring live running against a database and modifying the schema or access mechanisms to increase the performance of such systems
iv. Data archiving – establishing a strategy for archiving of ‘dead’ data
v. Data backup and recovery – establishing a procedure for backing-up data and recovering data in the event of hardware or software failure
Administration of the DBMS
The DBA would normally be expected to engage in the following key activities in relation to administering a DBMS:
i. Installation – taking key responsibility for installing DBMS or DBMS components
ii. Configuration control – enforcing policies and procedures for managing updates and changes to the software of the database system
iii. Monitoring DBMS usage and tuning DBMS – monitoring live running of database systems and tailoring elements of the DBMS structure to ensure the effective performance of such systems
Administration of the Database Environment
By administering the database environment, we mean monitoring and controlling the access to the database and DBMS by users and application systems.
Activities in this area include:
i. Data control – establishing user groups, assigning passwords, granting access to DBMS facilities, granting access to databases
ii. Impact assessment – assessing the impact of any changes in the use of data held within database systems
iii. Privacy, security and integrity – ensuring that the strategies laid down by data administration for data integrity, security and privacy are adhered to at the physical level
iv. Training – holding responsibility for the education and training of users in the principles and policies of database use