An oft-repeated story about database administration underscores both the necessity for database administration and the lack of understanding of a DBA’s function. It goes something like this:
The CIO of Acme Corporation hires a management consulting company to streamline their information technology (IT) operations. The consultant, determined to understand the way Acme works, begins by interviewing the CIO. One of his first questions is: “So, I see that you have a DBA on staff. What does he do?”
The CIO replies, “Well, I’m told that we need the DBA to make sure our production databases stay online. I know that some of our critical business processes like order entry and inventory use Oracle, but I really don’t know what the DBA does. But please don’t tell me I need another one, because we can barely afford to pay the one we have!”
This is a sad, but too often true, commentary on the state of database administration in many organizations. DBMS software is so complex these days that very few people understand more than just the basics (like SQL). However, DBAs understand the complexities of the DBMS, making them a valuable resource. Indeed, sometimes the only source of database management and development knowledge within the organization is the DBA.
The DBA, often respected as a database guru, is just as frequently criticized as a curmudgeon with vast technical knowledge but limited people skills. Just about every database programmer has his or her favorite DBA story. You know, those anecdotes that begin with “I had a problem…” and end with “…and then he told me to stop bothering him and read the manual.” DBAs simply do not have a warm and fuzzy image. However, this perception probably has more to do with the nature and scope of the job than with anything else. The DBMS spans the enterprise, effectively placing the DBA on call for the applications of the entire organization.
The truth is, many database problems require periods of quiet reflection and analysis for the DBA to resolve. Therefore, DBAs generally do not like to be disturbed. However, due to the vast knowledge most DBAs possess (the guru, again), their quiet time is usually less than quiet; constant interruptions to answer questions and solve problems is a daily fact of life.
DBAs, more than most, need to acquire exceptional communication skills. Data is the lifeblood of computerized applications. Application programs are developed to read and write data, analyze data, move data, perform calculations using data, modify data, and so on. Without data, there would be nothing for the programs to do. The DBA is at the center of the development life cycle — ensuring that application programs have efficient, accurate access to the corporation’s data. As such, DBAs frequently interface with many different types of people: technicians, programmers, end users, customers, and executives. However, many DBAs are so caught up in the minutiae of the inner workings of the DBMS that they never develop the skills required to relate appropriately to their coworkers and customers.
So what is a DBA? The short answer is simple: A DBA is the information technician responsible for ensuring the ongoing operational functionality and efficiency of an organization’s databases and the applications that access those databases.
The long answer to that question requires a book to answer. But let’s start by endorsing the need to tackle database administration as a management discipline: not art, not magic, but a craft that ensures the ongoing operational functionality and efficiency of an organization’s databases and applications.
Database administration is rarely approached as a management discipline. The term discipline implies a plan, and implementation according to that plan. When database administration is treated as a management discipline, the treatment of data within your organization will improve. It is the difference between being reactive and proactive.
All too frequently, the DBA group is overwhelmed by requests and problems. This ensues for many reasons, such as understaffing, overcommitment to supporting new (and even legacy) application development projects, lack of repeatable processes, or lack of budget. The reactive DBA functions more like a firefighter than an administrator; he attempts to resolve problems only after problems occur. The reactive DBA is focused on resolving the biggest problem confronting him.
In contrast, the proactive DBA implements practices and procedures to avoid problems before they occur. A proactive database administrator develops and implements a strategic blueprint for deploying databases within the organization. This plan should address all phases of the application development life cycle. A data specialist, usually the DBA, should be involved during each phase of the cycle. During the initiation and requirements gathering phase, the DBA must be available to identify the data components of the project. He can help to determine if the required data already exists elsewhere in the organization or if the data is brand new. During the analysis and design phases, the rudimentary data requirements must be transformed into a conceptual and logical data model.
Before development can begin, the logical data model must be translated to a physical database design that can be implemented using a DBMS such as Informix, SQL Server, Oracle or DB2. Sample data must be populated into the physical database to allow application testing. Furthermore, the DBA must develop and implement a process to refresh test data to enable repeatable test runs.
When the application moves from development to operational status, the DBA ensures that the DBMS is prepared for the new workload. This preparation includes implementing appropriate security measures, measuring and modifying the storage and memory requirements for the new application, and anticipating the impact of the new workload on existing databases and applications. The DBA is also responsible for migrating the new database from the test environment to the production environment.
While the application is operational, the DBA performs a host of duties including assuring availability, performance monitoring, tuning, backup and recovery, and authorization management. However, no application or database remains static for long. Because business needs change over time, the IT systems that support the business will also change. When maintenance is requested, the DBA becomes engaged in the entire process once again, from requirements gathering to implementation.
Finally, when the application reaches the end of its useful life, the DBA must help to determine the final status of the data used by the application. Is the data no longer required, or do other applications and processes use the data, too? Are there regulations that require the data to be stored longer than the life of the application? Does the business have any privacy policies that impose special rules for handling the data?
The DBA is responsible for managing the overall database environment. Often this includes installing the DBMS and setting up the IT infrastructure to allow applications to access databases. These tasks need to be completed before any application programs can be implemented. Furthermore, ad hoc database access is a requirement for many organizations.
Additionally, the DBA is in charge of setting up an ad hoc query environment that includes evaluating and implementing query and reporting tools, establishing policies and procedures to ensure efficient ad hoc queries, and monitoring and tuning ad hoc SQL.
As you can see, a good DBA is integral to the entire application development life cycle. The DBA is in demand for his knowledge of data and the way in which data is managed by modern applications.
This blog was originally published on Craig Mullins’ blog at https://datatechnologytoday.wordpress.com/2011/02/27/dba-as-a-management-discipline/
EPM applications help measure the business performance. This post will help you choose the best EPM solutions for your organization’s needs and objectives.
Which RAID should you use with SQL Server? Learn the differences between RAID 0, RAID 1, RAID 5, and RAID 10, along with best practices.