The Correlation Between Performance and Costs

Bookmark and Share

When you think about the role of a database professional, you probably don’t include “cost savings” in the list of responsibilities. In fact, you probably sum up the role with something along the lines of “keeping applications running smoothly.”

While that is a critical responsibility—especially given how important applications are to modern business and, in turn, how important databases are for applications—by and large, databases are misunderstood and underappreciated. And often, so are database professionals.

Maybe if there were a clearer correlation between the work of database professionals and money, people would pay more attention. Well, it turns out, there is.

The Financial Impact of DBAs and Databases

Generally speaking, there are three ways database professionals can have a direct financial impact:

1. Proactive database management prevents downtime: If the database stops functioning, everything stops working. A slow database decreases productivity for every user of an application. How many times have you heard a bank teller, hotel reception clerk, or any other customer service representative say, “Sorry, our computers are a little slow today”?

2. Moving workloads to lower cost alternatives: Database professionals can help transition applications from high-license-cost DBMS systems to lower-cost or even open source systems. Even though license costs are only a percentage of total costs, for certain applications, moving from an “enterprise” database to MySQL or PostgreSQL can result in significant savings.

3. Improving database performance: Yes, database professionals can also have an impact on the financial health of a company through improving database performance. There is a direct connection between the two, even if far too many are not yet aware of it.

The first two here should not come as a surprise, but the third is much less well-understood. Let’s discuss it further.

Improving Database Performance to Reduce Costs

For most SaaS businesses, as an example, a significant percentage of their expenses is the cost of infrastructure technology—most often in the cloud. We also know that the most critical component of an application is the database. Of course, it’s also the least understood.

When an application needs more performance, it’s easy to scale in the cloud by adding more hardware. If a database is struggling to meet end-user expectations, too often, the default reaction is to provision larger, faster hardware. You can add memory, move to a database instance with higher CPU capacity, pay for provisioned input/output operations per second (IOPS), or move to solid-state disk storage (SSD). In the cloud, each one of these results in a larger bill at the end of the month.

Unfortunately, all too often, these decisions are made without specific knowledge of whether or how they will help with database performance because few people really understand what happens inside a database and the specific bottlenecks that may be holding performance back. The result is continuously increasing infrastructure spending. Most companies accept these increasing costs as a fact of running a technology-based business.

In these situations, a database professional can be the performance guru and the person who changes the financial profile of the organization. With the right tools and information, he or she can see what happens inside the database to tune queries, remove bottlenecks, and identify exactly what is slowing down an application. The result is a more efficient application that consumes less infrastructure resources and therefore costs less to operate, and is a more intelligent investment overall.

A company with such an approach only adds CPU when they know CPU is the bottleneck, and they know the exact performance impact they will see with the additional investment. Likewise, they only use SSD drives when they know storage read/write is a significant contributor to performance. And so on and so forth.

Such a company has achieved performance certainty, meaning they know how their system performs, why it performs that way, what the drivers of performance are, and that the system has been optimized to run at top speed while consuming the minimum resources needed.

As you can see, in the cloud there is a clear connection between performance and resources, and between resources and costs. Well, you know what? The same connection exists on-premises, but the way companies operate often obscures the correlation. However, that doesn’t mean these opportunities to drive cost savings and make resources available for projects that truly need them don’t exist. It just takes a proactive database professional to find them and to educate management of their value.