☰ Chapters
Week 7 · Database Administration
CT004-3.5-3 Advanced Database Systems · Week 7

Database Administration

Technological, managerial and cultural aspects; the Data Administrator (DA) vs the Database Administrator (DBA).

Contents
    Learning outcomes

    Why database administration matters

    A DBMS is just a tool for managing data; it must be used effectively to produce the desired results. Introducing a DBMS is a big change and can have a big impact — positive or negative, depending on how it is administered.

    The process involves three important aspects:

    AspectConcerned with
    TechnologicalDBMS software and hardware
    ManagerialAdministrative functions
    CulturalCorporate resistance to change

    Technological aspect

    This aspect has major financial impact: the choice of database platform should be planned ahead for at least ~5 years.

    Managerial aspect

    Cultural impact

    A DBMS is likely to affect people, functions and interactions:

    Individual department managers must surrender ownership of “their” data to the data administration function and share it with the rest of the company. People may be unwilling to use the information provided and may question its accuracy. The database administration department must listen to end users’ concerns and act by educating them about the benefits.

    From the lecture notes · Lewin’s change model

    Organizational culture is the shared values, beliefs, customs and written/unwritten rules that shape how an organization behaves. Introducing a DBMS means managing change through three stages:

    1. Unfreeze — people are comfortable with the status quo; significant effort (“push” methods, then “pull”) is needed to get them ready for change.
    2. Transition — change is a journey, not a single step. It takes time, leadership, coaching and support; leaders must not expect everyone to cross in one bound.
    3. Refreeze — establish the new way of working as the stable norm.

    Evolution of the database administration function

    Departmentaldata processing CentralisedEDP Formal informationsystems department file-based “islands of information” → shared database→ need for central data management

    Data administration vs database administration

    Many large organizations have two main roles for managing company data:

    Data Administration (DA)

    A high-level function responsible for the overall management of data resources in an organization, including maintaining corporate-wide definitions and standards.

    Database Administration (DBA)

    A technical function responsible for technical issues such as security enforcement, database performance, and backup and recovery.

    In a small organization, the same person often does both.

    Data Administrator (DA) responsibilities

    Database Administrator (DBA)

    DBA’s responsibilities

    Coordinating, monitoring and allocating database administration resources — in eight areas:

    1. Selection of DBMS, utilities and hardware
    2. End-user support
    3. Enforcing policies, procedures and standards
    4. Data security, privacy and integrity
    5. Data backup and recovery
    6. Data distribution and use
    7. Design and implementation of databases and applications
    8. Authorisation management

    1. Selection of DBMS, utilities and hardware

    An important technical responsibility. Factors to consider:

    DBMS modelStorage capacityApplication development supportSecurity & integrityBackup & recoveryConcurrency controlPerformanceDBA toolsData distributionPortability & standardsHardwareVendor training & supportThird-party toolsCost

    e.g. development support (does it work with PHP/Java?), performance (Access vs SQL Server), hardware (servers, terminals, network).

    2. End-user support

    3. Policies, procedures and standards

    Continuous enforcement of:

    TermMeaningExample
    PoliciesGeneral statements of direction that support goals“All customer data must be backed up.”
    StandardsMore detailed and specific than policies; describe minimum requirements“Passwords ≥ 12 characters; table names singular, PascalCase.”
    ProceduresWritten instructions describing a series of steps to be followed“Steps to create a new user account.”
    Key distinction

    The DA sets policies, procedures and standards; the DBA enforces them.

    4. Data security, privacy and integrity

    5. Data backup and recovery

    6. Data distribution and use

    7. Design and implementation of databases and applications

    8. Authorisation management

    AreaActivities
    User access managementDefine each user; assign passwords; define user groups; assign access privileges; physical access control
    View definitionDefine views to protect and control the scope of data accessible to users
    DBMS utilities access controlPlace limits on the use of query and reporting tools
    DBMS usage monitoringAudit the use of data in the database; create an audit log

    Contrasting DA and DBA

    Data Administrator (DA)Database Administrator (DBA)
    Strategic planningControl and supervision
    Sets long-term goalsExecutes plans to reach goals
    Sets policies and standardsEnforces policies and procedures; enforces programming standards
    Broad scopeNarrow scope
    Long termShort-term focus on daily operations
    Managerial orientationTechnical orientation
    DBMS independentDBMS specific
    Conceptual & logical designPhysical design
    Exam tip

    For “compare DA and DBA” questions, give each difference as a pair (strategic vs control, sets vs enforces, broad vs narrow, long-term vs daily, managerial vs technical, DBMS-independent vs DBMS-specific). A table answer is ideal.

    Quick review

    What are the three aspects of introducing a DBMS?
    Technological (software/hardware), managerial (administrative functions), cultural (resistance to change).
    Why must department managers “surrender ownership” of data?
    Data becomes a shared corporate resource managed centrally by data administration, so it can be integrated and shared across the company.
    List the eight DBA responsibility areas.
    Selection of DBMS/utilities/hardware; end-user support; policies, procedures & standards; security, privacy & integrity; backup & recovery; data distribution & use; design & implementation; authorisation management.
    Policies vs standards vs procedures?
    Policies: general direction. Standards: detailed minimum requirements. Procedures: written step-by-step instructions.