Traditional mainframe roles are undergoing a metamorphosis that transforms the emphasis from product-centric specialists to a multi-role generalist. With budget cutting and the mandate to “do more with less,” this transformation extends to Database Administrators (DBAs) tasked with multiple responsibilities. Often, DBAs must shift from generalist to specialist and back, and always in an increasingly dynamic, complex environment.
But has technology kept pace and helped DBAs become more effective in managing their many responsibilities? Not really. While enhancements to operating system components, software products, and hardware innovations such as the IBM mainframe specialty processors have helped achieve business objectives and improve customer service, little has been done to help improve the technologist’s ability to enhance their effectiveness.
DBAs are under relentless pressure to speed and automate their tasks, ensure they’re visualizing the “big picture” of their infrastructure, and manage change at a moment’s notice, all while ensuring data integrity and maintaining business continuity and Service Level Agreements (SLAs). Future technology must provide a platform for a unified, radical change that will help DBAs gain greater control of their DB2 environment, increase productivity, and improve the accuracy and delivery of database services enterprisewide.
Technology needs to reflect the realities of data center life, improving the skills and capabilities of both experienced and novice DBA technologists. Future innovations must help DBAs:
- Become more effective at many roles, previously performed by multiple or even groups of people
- Be agile and scalable, instantly apply best practices to handle complex workloads and any related customizations
- Quickly ramp up novice staff to perform tasks done now by people with 30 years of experience
- Transparently comply with internal and external mandates that require segregation of duties to help mitigate risk and exposure.
Imagine an environment where all the disparate functions required by a DBA for specific tasks are provided in a consistent way, where knowledge is presented simply and the output from multiple tools can be used as input to other tools. This role-based capability would radically enhance the overall user experience to give all levels and roles of users, including expert and novice technicians, a much better work environment and vast efficiency improvements.
The first improvement would be quantum-leap changes in user workspace ergonomics, something beyond mere “dashboard” visual reporting. Sure, it should be beautiful to look at and use, but this customizable workspace should also make you feel more energetic and productive.
DBAs are constantly interrupted during their daily work. People walk up and ask for help, the phone rings, or an email alert arrives that requires immediate attention. Today, it isn’t unusual for a DBA to have up to 15 or more windows open to work on different aspects of a problem. In the future, a single workspace would improve the user experience by unifying data, performance and system health, as well as provide simple navigation into tasks, objects, session history and more.
This role-based workspace would be able to quickly and automatically chart, trend, and graph thousands of data points, including data, performance metrics, and time ranges vs. pouring over tabular text data (i.e., extract performance data and import into a common spreadsheet). The user would see collaborative notes of important events posted by the system or other DBAs, and have the ability to add notes and comments themselves, so the full spectrum of change can be visualized against performance.
The next area for improvement would be to help DBAs dramatically reduce the time required to solve problems. DBAs tend to spend lots of time doing investigative analysis. It all starts with an issue. A DBA might receive a notification via email, a phone call, a ticket through the problem reporting system, or during the course of monitoring a particular process. Depending on their experience level and tools available, they may or may not know where to begin the investigation.
For example, a DBA might need to analyze the SQL statement access path report generated through the EXPLAIN function. This task can be complex and requires a high degree of expertise. A trained DBA can use this report to make recommendations on what needs to be changed in either the SQL statement or the DB2 object to correct the problem. However, this doesn’t account for workload history and what’s “normal” for this SQL statement. To figure that out, the DBA must have immediate access to current information and system status such as: