The Siloed Database Administrator: Breaking down the wall between DBAs and the rest of IT

Featured article by Gerardo Dada, vice president of product marketing and strategy, SolarWinds

It’s a rare IT department that doesn’t experience a little bit of the “blame game” when something goes awry—particularly in the first moments after an application goes down or Help Desk tickets start flooding in.

The systems team may point immediately to the network folks, or perhaps everybody is looking at the database administrator (DBA). However, this finger pointing does little to actually solve problems, and can serve as an impediment to getting to the root of a problem, which only exacerbates already-critical IT performance issues.

The root cause in most cases is lack of visibility. Each team is looking at its own dashboard focused on a specific element of the infrastructure. It’s becoming more widely accepted that as infrastructures increase in complexity, IT can no longer function in these vacuumed silos, regardless of size or industry.

However, one particular role—that of the DBA—appears to be holding fast to its own island. To what degree is this perception is true, and how (and why) should organizations break down the wall between the DBA and the rest of IT?

Application Performance Matters, Which Means the Database Matters

Today’s businesses operate in an increasingly application-centric world where end users rely on high-performing, always available applications—from email and collaboration to CRM and finance. The high performance expectations end users have for these applications is largely driven by consumer-oriented experiences with sites and services such as Google and Facebook. In fact, response time expectations for websites has been shown to increase every year. These expectations transfer to workplace applications. End user perception of application performance has changed to the point a “slow” application may as well be considered a “broken” application.

In fact, a recent SolarWinds survey found that 93 percent of business end users say application performance and availability affect their ability to do their job, with 62 percent saying it is absolutely critical.

Furthermore, another recent SolarWinds survey showed just how important the database is to maintaining optimal application performance. For example, consider the following findings:

– 70 percent of IT pros agree that application performance is contingent on database performance

– 85 percent of IT pros cited database performance as challenge they want to solve

The Role of the DBA in Maintaining Application Performance

With the data above, a connection can easily be made between maintaining optimal application performance to the role of the DBA—essentially, DBAs are in a position to be the “performance gurus” within IT, and they have arguably the best information about how systems are performing. This in and of itself is reason to ensure the DBA is well-integrated among the larger IT organization.

However, the DBA often has his or her own tools and processes, without much collaboration with other teams. Part of this is due to the sheer complexity and mission-critical nature of the database. Additionally, there may not have been enough of an incentive until now to have the DBA effectively brought into the fold; perhaps due to resistance from the DBA or the CIO failing to see the need.

Paving the Way for a Less-Siloed DBA

However, there are trends and innovations leading toward a less-siloed DBA—and IT department in general. Here are two examples:

DevOps: This software development method, which stresses communication between development and operations, is gaining popularity as a way to streamline IT and can also be looked at through the lens of development, DBAs and operations teams. Operations, which can lack insight into development, can be quick to blame code. Meanwhile, developers don’t have a view into production and blame the database or the infrastructure. Using DevOps best practices to enhance communication between developers and DBAs, IT should consider:

1. Providing developers with direct monitoring visibility into test, staging and production servers so they can understand the impact of the code they are producing

2. Building performance into the development process by making it a functional requirement expressed in end user response times

3. Establishing share metrics and a basis for equal access to metric reports across teams

4. Understanding the performance impact of each element of the application stack to identify bottleneck and the breakdown of what elements have a bigger role in performance

5. Adopting an agile mentality that prioritizes adaptive planning, evolutionary development, early delivery, continuous improvement and rapid/flexible response to change

Performance Management: Whereas IT departments formerly prioritized ensuring systems and networks were simply up and running, changing end user expectations as described above are shifting the focus to performance management as a competency. The bar from getting technology infrastructure to “just work” to now having it work optimally is set and there’s no going back. With the growing sophistication of affordable monitoring and management tools, IT can now access, analyze and share powerful and easy-to-digest IT metrics among the department and with company stakeholders, further driving the ability to emphasize performance. Building a performance management competency requires:

1. Recognizing that performance, and not just uptime, are important metrics for IT

2. Making end user response time a shared goal for the entire organization

3. Providing visibility to the entire team and pulling in the DBA to effectively prioritize performance.

4. Building the organization’s knowledge and toolset for effective performance management, thereby enabling the business to move at the speed of IT

The Role of the DBA in Breaking Down IT Siloes

DBAs can also take an active role in breaking down silos themselves by taking the following steps:

1. Share data and information freely to show other departments how the database side of the house works; metrics and analytics communicate trends and issues in a neutral, impactful way that can drive effective conversation and help uncover solutions.

2. Avoid the blame game when possible. The DBA can aim to set an example and proactively get to the root of a problem—backed up by data collected from monitoring and management tools—to diffuse a tense situation while displaying a spirit of collaboration.

3. Initiate follow-up after an issue’s root cause has been identified and the problem has been resolved to discuss how IT as a group can avoid it from happening in the future.

4. Own up when the issue does arise from the database—it happens, and removing the “air of mystery” will build good faith with colleagues when it’s not actually the database.

5. Consider regular cross-team check-ins, either together or separately, with systems, network, security, development, virtualization and other teams. Discuss new projects or simply be available to answer questions they may have about the database.

While the DBA does bear his or her share of abuse when something goes awry, IT systems do not work in isolation. Most problems are often caused somewhere between the interaction between elements. By focusing on removing the DBA silo within the organization, IT can start to move beyond the blame game and focus on driving IT performance—and the needs of the business.

Gerardo Dada is Vice President of Product Marketing and Strategy for SolarWinds’ database and applications business globally, including SolarWinds Database Performance Analyzer. Gerardo is a technologist who has been at the center of the Web, mobile, social and cloud revolutions at companies like Rackspace, Microsoft, Motorola, Vignette and Bazaarvoice. He has been involved with database technologies from dBase and BTrieve to SQL Server, NoSQL and DBaaS in the cloud.