The Next Generation DBA – Introducing the Database Concierge

Last month, I had the honor of delivering a keynote address at the GuideShare Europe UK Conference to discuss the dramatic changes in the priorities and demands of the next generation of database administrators (DBAs). I want to thank GSE UK for having me, and all who attended.

For those who were not able to attend, or anyone that would like to revisit the topics discussed, I will be highlighting some of the major themes and insights from my presentation here in a blog series.

The Realities of Today’s Enterprise

To survive in today's digital economy, organizations must capitalize on their data growth and manage it effectively, which is increasingly challenging with data distributed across clouds, applications and machines. While databases are getting larger and the availability requirements longer, DBA teams are shrinking for these main reasons:

• The increase in available and inexpensive storage has shifted the focus of the job away from storage management, resulting in a reduction of staff
• Attrition due to retirements in an aging workforce
• Most new hires are not acquiring z-based skills at the university level
• The traditional role of the DBA is being increasingly automated or outsourced

To meet the needs of the business, the DBA needs concierge-level assistance to maintain peak effectiveness in expected areas, while expanding for new business demands. The goal of this new solution, and what I would call the “Database Concierge” has a few key purposes:

• Enabling self-service to all IT professionals
• Providing remaining DBAs specific advice using deep understanding of z/OS
• Building the new generation of experts

IT is a part of what drives enterprises and the products and services they deliver to customers, so technology will continue to redefine the relationship between customer and business. Organizations need to prioritize their investment in current technologies to enable profitable growth and future agility, as well as to substantially reduce the cost of operations through automation.

Bottom line: DBAs need the freedom to provide more value to the business, meaning their software needs to become smarter and easier to use.

Think Optimization, but on Steroids

Large IT installations have a wealth of data about capacity and performance, but organizations often struggle to create value from this data. Successful optimization requires tools that create transparency – combined with people and processes focused on cost-hunting.

For instance, consider the challenges in tuning queries across the organization. Writing good code and writing optimized SQL requires different skills sets. Application developers rely on frameworks to write the SQL for them, if they focus on capability delivery, performance can become a lower priority. Often the QA environment is inadequate to shake out scalability issues, and while experienced DBAs have the skill set to optimize SQL, their numerous responsibilities and the disconnect between application and database groups means they often don’t get a chance to focus on SQL optimization until problems arise.

• According to IBM data, DBAs can spend 25% – 30% of their time managing app performance and tuning SQL queries.

Once in production, query tuning is incredibly challenging and performance problems can appear out of the blue. When the data changes (data volume, data distribution, etc.), access paths can change, too. For example, did the database design change, was an index added or dropped? Now, DBAs are responsible for things like Index Elimination strategies.

They need to get information from several places including the catalog performance monitors, EXPLAIN data, etc. and establish solid z/OS system monitoring to not only prevent any problems, but to augment their other methods of maintaining availability. DBAs require a deep understanding of database concepts to assure good database and query design -- in other words, they need a concierge to think like an optimizer.

Optimization Helps DBAs Create a Ripple Effect

There’s a synergistic partnership between the DBA and the concierge-level service for optimization. The solution eliminates guess work, enabling application outages and performance issues to be resolved faster and with greater confidence. The next-gen DBA could then quickly act and think like an expert, without feeling like they were in firefighting mode.

For example, a DBA in the banking industry with concierge-level software would create a ripple effect improving all other areas of the business, empowering the:

• Head of Retail Banking to deliver business results needed for growth
• Senior App Dev Executive to ensure high quality customer experiences for online banking
• IT Infrastructure Director to provide the application criteria needed by the business
• Lead Mainframe Software Developer to deliver performance to meet SLAs

AI/ML & The Next-Gen DBA

Optimizing intelligent software or systems is a time consuming and rare skill. In a growing organization, it becomes harder and harder just to manage alerts from performance monitors that were either set too low or too high, thus losing their validity. The next-gen DBA will need concierge-level software to apply AI to thresholds for anomaly detection to reduce their manual workload.

DBAs need a way to filter out data to remove the guesswork and focus on what is important while using the performance history to provide more value. With ML determining ‘normal’ over time, the DBA would no longer need to approximate or guess what a good threshold is for a given SQL statement. Therefore, alerts and exceptions would become much more meaningful. Utilizing AI, such solutions can make true self-healing, autonomous databases possible.

Today’s IT environments require skill sets that challenge even accomplished database administrators. Every customer has different utilization, meaning what is normal to one customer is abnormal to another. Who knows more about your environment or history than your performance monitor? In conjunction with AI, database concierge software can learn and adapt as your application changes.

Proactive DBAs Deliver More Value

Understanding and optimizing Db2 system resource usage requires decades of skill. Concierge-level software would learn patterns, reduce false positives, easily identify concerns with historical expert knowledge, predict future resource shortages and at the same time, allow the DBA to be proactive rather than reactive. DBAs must be positioned to deliver significant value to the organization to positively impact every other area of the business. As we consider how to empower the next generation of DBA professionals, optimizing their tools and workflows is a critical first step.

Learn more here, and be sure to check back on the blog for the next part in this series, focused on open-source practices for DBAs.