Professional Documents
Culture Documents
Contents: Introduction Why do we need Database Administrator for an organization? Roles of Database Administrator Planning for databases Managing data security Backup database Managing Data Quality Security policy and Disaster Recovery Functions of Database Administrator Skills and Activities of Database Administrator Database Administrator Activities and Services Conclusion Reference
Database Administrator
DEFINITION Data: Data are nothing facts, text that have meaning in the users environment. Database: An organized collection of logically related data. Metadata: Data that describes the properties or characteristics of
other data.
Database Administrator:
The person who is responsible for physical database design and for dealing with technical issues such as security enforcement, Database performance, Backup and recovery associated with managing a Database.
Database Administrator
The Database Administrator in a line position has both the responsibility and the authority to plan, define, implement, and enforce the policies, standards, and procedures used in the data administration activity.
Database Administrator
prototyping approaches, is also necessary. Strong design and data modeling skills are essential at the conceptual, logical, and physical levels. But managerial skills are also critical: the Database Administrator must manage other Information Systems (IS) personnel as the database is analyzed, designed, and implemented, and the Database Administrator must also interact with and provide support for end users who are involved with the design and use of the database.
Database Administrator
typical organization. It is critical that those involved in Database Administration understand the information requirements of the organization and be able to contribute to the development of the information architecture that will those needs. Data Conflict resolution Databases are intended to be shared, and usually involve data from several different departments of the organization. Ownership of data is a ticklish issue at least occasionally in every organization. Those in Database Administration are well placed to resolve data ownership issues because they are not typically associated with a certain department. Establishing procedures for resolving such conflicts is essential. If the administration functions have been given sufficient authority to mediate and enforce the resolution of the conflict, they may be very effective in this capacity. Internal Marketing While the importance of data and information to the organization has become widely recognized with in the organization, it is not necessarily true that an appreciation of data standards has also evolved. The importance of following established procedures and policies must be protectively instituted through data and Database Administrators. Effective internal marketing may reduce resistance to change in data ownership problems. Managing the information repository Repositories contain the metadata that described an organizations data and data processing resources. Information repositories are replacing data dictionaries in many organizations. While data dictionaries are simple data-element documentation tools, information repositories are used by data administrators and other information specialists to manage the total information-processing environment. The repository is an essential tool for Database Administration and is used through out entire database system life cycle. Selection of hardware and software The evolution of selection hardware and software is critical to an organizations success. The increasingly rapid development of new hardware and software and the rapid changes with in organizations creates a very demanding situation. A vendors database that is the most appropriate product available with in the choice of platform is made by turnout to be an unfortunate choice of a competitor releases a new version that has different and appropriate functionality. The current vendor may even go out of business, resulting in production database that is a nightmare to maintain. Database Administrators are being expected to know more about hardware architectures and to be able to administer both in-house-developed applications off-the-self applications, as organization strive to meet their information needs more rapidly.
5
Database Administrator
Installing and upgrading the DBMS Once the DBMS is selected, it must be installed. Before installation, benchmarks of the workload against the database and a computer supplied by the DBMS vendor should be run. Benchmarking anticipates issues that must be addressed during the actual installation. A DBMS installation can be complex process of making sure all the correct versions of different models are in place, all the proper devices derives and presents, and the DBMS works currently with any third-party software product. DMBS vendors periodically update package modules: planning for, testing, and installing upgrades to ensure existing applications still work properly can be time-consuming and intricate. Once the DBMS is installed, user accounts must be created and maintained. Tuning database performance Because databases are dynamic, it is improbable that the initial design of the database will be sufficient to achieve the best processing performance for the life of the database. The performance of the database needed to be constantly monitored. The design of the database must be frequently changed to meet new requirements and to overcome the degrading effects of many content updates. Periodically, the database must be rebuilt, reorganized, and reindexed to recover wasted spaces, to correct poor data allocations and fragmentation across storage devices, to allocate different buffer space for concurrent processing, and to reset parameters that define the best recovery procedures. Various system tables, needed by the DBMS but not containing business data, periodically need to be rebuilt to be consistent with the new size and use of the database. Improving database query processing performance The work load against database must certainly will extended over time has more users find more ways to use the growing amount of data in a database. Thus, some queries, which originally ran quickly against a small database, may need to be rewritten in a more efficient form to run in a satisfactory time against a fully populated database. Indexes may need to be added or deleted to balance performance across all queries. Data may need to be relocated to different devices to allow better concurrent processing of queries and updates. Arguably, the vast majority of the time spent by a Database Administrator will be for tuning database performance and improving database query processing time. Managing data security, privacy, and integrity Protecting the security, privacy, and integrity of the organizational database rests with the Database Administration function. Here it is important to realize that the advent of the internet and intranets to which databases are attached, along with the possibilities for distributing data and database to multiple sites, having complicated the management of data security, privacy and integrity.
6
Database Administrator
Data backup and recovery The Database Administrator must ensure that backup procedures or established that will allow the recovery of all necessary data, should a loss occur through application failure, hardware failure, physical or electrical disaster, a human error or malfeasance. These strategies must be fulfilled and evaluated at the regular intervals. Planning for Databases The initial phase of the database development life cycle is planning. It is at this point that a purposed system should be carefully considered for the view point of the enterprise architecture. Database design occurs with in the constraints of the enterprise information system. During the planning phase an initial assessment of the proposed database system is made. The proposed system should contribute to meeting the businesss goals and strategies, which have been developed has part of the enterprise architecture. The other components of the scope layer in enterprise architecture are also important. The initial assessment will also include consideration of continuing users of the existing system, modifying the existing system, or replacing the existing system. The decision will rest with the extent to which flaws in the current system exist, in the possibilities of correcting those flaws. If a new system is a possibility, then a feasibility study should be conducted. The feasibility study should address of the technical accept of the project, including hardware and software purchases or development cost, operating environment, system size and complexity, and level of experience with similar systems. At this point this is not necessary to identify particular vendors or products, but classes of implementation, such as PC or Mainframe hardware, database model, and programming language should be consider. The feasibility study should also address the estimated cost of the project, both tangible and intangible. Costs should be identified as one-time costs associated with the project are recurring costs that will contain through out the life of the system. Operational feasibility, an assessment of likelihood that the project will meet its expected objectives, should also be determined. The likelihood of completing the project with in the expected time schedule should also be estimated. Any potential legal are contractual problems should be identified now. And, political implications of the project should be considered. Key stake holders support for the project will greatly improve its prospect of success. Once the feasibility information has been collected, the formal review of the project that includes all interested parties can be conducted. The focus of this review is to verify all information and assumptions in the base line plan that has been developed before moving a head with the project. After the project is approved, problems defined during the planning phase will be studied further during the analysis phase of the project. Managing data security
7
Database Administrator
The goal of database security is protection of data from accidental or international threats to its integrity and access. The database environment has grown more complex, with distributed database located on client/server architecture rather than the mainframes. Access to data has became more open through the internet. As the result, managing data security effectively has became more difficult and the time consuming. Database Administrator is responsible for developing overall policies and procedures to protect the database. Threats to Data Security Threats to data security may be direct threat to the database. For example, those who gain unauthorized access to a database may then browse, change, or even steal the data to which they have gained access. Focusing on database security alone, however, will not ensure a secure database. All parts of the system must be secure, including the database. The network, the operating system, the buildings in which the database resides physically, and the personnel who have any opportunity to access the system. The following threats must be addressed in a comprehensive data security plan: Accidental losses, including human errors, software, and hardware-caused branches Establishing operating procedures such as user authorization, uniform software installation procedures, and hardware maintenance schedules are examples of actions that may be taken to address threats from accidental losses. As in any effort that involves human beings, some losses are inventible, but will through out policies and procedures should reduce the amount and severity of losses. Of potentially more serious consequence are the threats that are not accidental. Theft and fraud These activities are going to be perpetrated by people, quite possible through electronic means, and may or may not alter data. Attention here should focus on each possible location. For example control of physical security, so that unauthorized personnel are not able to gain access to the machine room, should be established. Data access policies that restrict alternating data immediately to a payroll run will help to secure the data. Establishment of a firewall to protect unauthorized access to inappropriate parts of the database through outside the communication links in another example of the security procedures that will harm for people who are intend on thefts or frauds. Loss of data integrity
8
Database Administrator
When data integrity is compromised, data will be invalid or corrupted. Unless data integrity can be restored to establish backup and recovery procedures. An organization may suffer serious losses or make incorrect expensive decisions based on invalid data. Loss of availability Sabotage of hardware, networks, or applications may cause data to become unavailable to users, which again may lead to serve the operational difficulties. Loss of privacy or confidentiality Loss of privacy is usually taken to mean loss of protection of data about individuals, while loss of confidentiality is usually taken to mean loss of protection of a critical organizational data, which may have strategic value of organization.
Backup Database Database recovery: Mechanisms for restoring a Database quality and accurately
after loss or damage. Database recovery is Data Administrators response. Inevitably, Databases are damaged or loss because of sum system problem that may be caused by human errors, hardware failure, incorrect or invalid data, program errors, computer virus, network failures, conflicting transaction, natural calamities. Since the organization depends heavily on its Database, the Database management system must provide mechanism for restarting a Database quality and accurately after loss or damage. Basic recovery facilities: A Database management system should provide four basic facilities for backup and recovery of a Database: 1. Backup facilities, which provide periodic backup copies of portion of or the entire Database. 2. Journalizing facilities, which maintain an audit trail of transaction and Database changes. 3. A checkpoint facility, by which the DBMS periodically suspense all processing and synchronizes its files and journals. 4. A recovery manager, which allows the DBMS to restore the Database to a correct condition and reset processing transaction.
Backup facilities
The DBMS should provide backup facilities that produce a backup copy of the entire Database plus control file and journals. Typically a backup copy is produced at least one per
9
Database Administrator
day. The copy should be stored in a secured location where it is protected from loss or damage the backup copy is used to restore the Database in the event of the hardware failure or damage. Some DBMS provide backup utilities for the Database Administrator to use to make backups: other systems assume the Database Administrator will use the operating system commands, expert commands, or SELECT..INTO SQL commands to perform backups. Since performing the nightly backup for a particular Database is repetitive, creating a script and automates regular backups will save time and result in fewer backup errors. With large Databases, regular backups may be impractical, as the time required to perform the backup by exceed that available. Or, a Database may be a critical system that must always remain available: so, a cold backup where the Database is shutdown is not practical. As a result, backups may be taken of dynamic data regularly. But backups of static data, which doesnt change frequently, may be taken less often. Incremental backups which record changes made since the last full backup, but which do not take so much time to complete, may also be taken on an interim basis, allowing for longer periods of time between full backups. Determining backup strategies, thus, must be based on the demands being placed on Database system. Journalizing facilities A DBMS must provide journalizing facilities to produce an audit trail of transaction and database changes in the event a failure, a consistent Database state can be reestablished using the information in the journals through with the most recent completely backup. The first is the transaction log, which contains a record of the essential data for each transaction that is processed against the Database. Data that are typically recorded for each transaction include the transaction code or identification action or type of transaction, time of the transaction, terminal number or user id, input data values, table and records accessed, record modification, and possibly the old and new file values. The second kind of log is the Database change log, which contain before-and afterimages of records that have been modified by transactions a before image is simply a copy of a record before it has been modified, and an after-image is a copy of the same record after it has be modified.
Checkpoint facility Checkpoint facility: A facility by which the DBMS periodically refuses to
accept any new transactions the system is in a quiet state, and the Database and transaction logs are synchronized. A check point facility in a DBMS periodically refuses to accept any new transaction. All transactions in progress are complemented, and the journal files are bought up to date. At this point the system is a quiet state, and the Database and transaction locks are synchronized. The DBMS writes a special record to the lock file, which is like a
10
Database Administrator
snapshot of the state of the Database. The check point record contains information necessary to restart the system. Any dirty data blocks are written from memory to disk storage, thus ensuring that all changes made prior to take the check point have been written to long term storage. A DBMS may perform check points automatically or in response to commands in user application program check pints should be taken frequently. When failures do occur, it is often possible to resume processing from the most recent check point. Thus, only a few minutes of processing work must be repeated, compared with several hours for reset of the days processing.
Recovery Manager: A module of the DBMS which restores the Database to a
correct condition when a failure occurs and which resumes processing user request. The type of restart used depending on the nature of the failure, the recovery manager uses the logs to restore the Database. Managing Data Quality Organizations must strive to identify the data that are relevant to their decision making, to develop business policies and practices that ensure the accuracy and completeness of the data, and to facilitate enterprise-wide data sharing. Establishment of a business information advisory committee consisting of representatives from each major business unit who have the authority to make business policy decisions can contribute to the establishment of high data quality. These committee members acts as liaisons between IT and their business unit, and consider not only their functional units data needs but also enterprise-wide data needs. The members need to have a strong interest in managing information as a corporate resource, an in-depth understanding of the business of the organization, and good negotiation skills. Such members are sometimes referred to as data stewards; people who have the responsibility to ensure that organizational application properly support the organizations enterprise goals. They must also ensure that the data that are captured are accurate and consistent throughout the organization, so that users throughout the organization can rely on the data. Achieving high speeds of data entry may override concerns about accuracy of data entry. Any organization needs to address five areas with regards to achieving high data quality: 1. 2. 3. 4. 5. Security policy and disaster recovery Personal controls Physical access controls Maintenance controls Date protection and privacy
11
Database Administrator
Security Policy and Disaster Recovery Every organization should establish comprehensive security policies and make detailed disaster recovery plans. The security policies will determine how the organization is going to maintain a secure system. Disaster recovery plans will determine how the organization will continue to function should an emergency situation, such as a flood or fire, arise. Written security policies should be established that provide guidelines where security measures will be necessary. Responsibilities and duties of the employees with regard to those policies should be established, and the consequences should the policies not r54r546be followed stated. Last, general procedures to be taken to implement the guidelines should be stated. These general procedures will be implemented through more specific procedures, which are the most likely part of the security policy to change over time, as new uses for technology develop and new products for enforcing a secure environment are developed. Disaster recovery plans should be worked out in detail prior to the occurrence of any disaster that may be foreseen, and if possible, tested. Contingency plans must consider data, equipment, personnel, and management issues involved in transferring operations to another location and safeguarding operations. Copies of plans should be kept in multiple locations and reviewed periodically. Personal Controls Adequate controls of personnel must be developed and followed, for the greatest threat to business security is often internal rather than external. Organizations should develop procedures to ensure a selective hiring process that validates potential employees representation about their backgrounds and capabilities. Employees should be trained in those aspects of security and quality that are relevant to their jobs, and be encouraged to be aware of and follow standard security and data quality measures. Standard job controls such as separating duties so no one employee has responsibility for an entire business process, or keeping application developers from having access to production systems, should also be enforced. Functions of Database Administrator 1. Database planning including the definition of standards, procedures, and enforcement. 2. Database requirements-gathering and conceptual design. 3. Database logical design and transaction design. 4. Database physical design and implementation. 5. Database testing and debugging. 6. Database operations and maintenance, including installing, conversion, and migration.
12
Database Administrator 7.
Database planning: Improve selection of technology by selectively evaluating possible products. Be sure to consider each technologys fit with the enterprise data model, seeking to find ways to reduce time required in later stages as a result of careful selection of technology at the Database planning stage. Database analysis: Try to work on physical design in parallel with development of the logical and physical models. Prototyping the application now may well lead to changes in the logical and physical data models earlier in the development process. Database design: Prioritize application transactions by volume, importance and complexity. These transactions are going to be most critical to the application. Specifications for them should be reviewed as quickly as the transactions are developed. Logical data modeling, physical Database modeling, and prototyping may occur in parallel. Database Administrators should strive to provide adequate control of the database environment while allowing the developers space and opportunity to experiment. Database implementation: Institute Database change-control procedures so that development and implementation are supported rather than slowed. Wherever possible, segment the model in to modules that can be analyzed and implemented more quickly. Find ways to test the system more quickly without compromising quality. Testing may be moved earlier in the development; use testing and change control tools to build and manage the test and production environments. Operation and maintenance: Review all timesaving measures that have been taken to ensure that Database quality has not been compromised. Consider using third-party tools and utilities wherever possible to save work; other tools such as Lotus Notes may reduce need for meetings, thus saving time.
Database Administrator
personnel management and on the interactions with the end user community. The Database Administrators technical role covers activities involving the DBMS database design, and use of application programs. The managerial skills are: Broad business understanding. Coordination skills Analytical skills Conflict resolution skills: oral and written Negotiation skills Experience: 2-5 years in large scale. The technical skills are: o Broad data processing background o Systems Development Life Cycle o Structured methodologies o Database Life Cycle o Database design and modeling skills 1. Conceptual (E-R) 2. Logical 3. Physical o Data Dictionary Management Database Administrator Activities and Services Support of the end user community Organizing of policies, procedures and standards Testing of data security, privacy, and integrity Monitoring of data backup and recovery Delivering of data distribution and use. Conclusion: The Database Administrator is responsible for the management of the corporate database. The internal organization of the Database Administrator function varies from company to company. Database Administrator has both technical and managerial responsibilities. The Database Administrator must have a very diverse mix of skills. The Database Administrator role as becoming more specialized, evolving into specialties. Database Administrators who gain broad experience and develop the ability to adapt quickly to changing environments will have many opportunities it is possible that some current Database Administrator activities, such as tuning, will be replaced by decision support systems. To help translate strategic plans into operational plans, the Database
14
Database Administrator
Administrator has access to an arsenal of database administration tools. Such as data dictionary and computer aided software engineering (CASE);
15