Inside SelectStar Expert Recommendations
These days database expertise is a moving target. The rapid expansion of data platforms has made it hard for anyone–even those dedicated to databases–to keep up. What’s more, increasing DevOps and BI adoption, a growing reliance on cloud database platforms and the evolving relationship between applications and databases means that more of the front-line database troubleshooting and performance tuning might be done by an application developer, rather than someone with a database background.
Figure 1: 5-year growth of non-traditional DBMS also illustrates how many NoSQL and NewSQL technologies have sprouted up (source: DB-Engines)
For these “DevBAs” as we call them, SelectStar expert recommendations are powerful tools to improve database security, boost performance and reinforce configuration standards. They provide specific advice on how to address key database and infrastructure best practices. In general, recommendations are more advanced than alerts because they often synthesize the best practices of the specific database platform you’re running–like MongoDB, PostgreSQL and Cassandra–along with the best practice information specific to the infrastructure you’re running that platform on–like Azure, AWS, VMware or the Google Cloud Platform. This combination provides vital information for tuning that cannot be procured from any other source.
Figure 2: View all your recommendations from the blue button on the main navigation.
Who are the “experts”?
Even if it’s your first time working with a particular DBMS, SelectStar expert recommendations bring loads of experience to the table. Our recommendations are based on:
- A panel of database administrators, each masters in a specific platform, who regularly review and revise our recommendations
- Documented best-practices from the data platform or infrastructure developers
- Real-world learnings from composite SelectStar customer data
Let’s breakdown a specific recommendation to demonstrate how it considers both the infrastructure and the platform best practices. For example, below you’ll see an out-of-the box recommendation to “Disable public accessibility to improve security.” This is based on a documented Amazon best practice, and is relevant to this PostgreSQL database’s security because it’s running on Amazon RDS. It’s easy to see how this recommendation might help a developer who might have missed a specific setting when he or she deployed this database instance. It also shows how a more central DBA team or ops admin could use recommendations to maintain consistent configuration standards across an enterprise, and how they could use recommendations to tune up after the database configuration has drifted.
Figure 3: Recommendations give you all you need to know with detailed descriptions, reasons and how to resolve the issue.
Customizing and creating recommendations
Any out-of-the-box SelectStar recommendation can be modified. So if you decide that a particular recommendation is not appropriate for your environment, you can turn it off. You can also adjust the settings or you can create my own recommendations based on your internal best practices.
Figure 4: Customizing a SelectStar recommendation takes just a couple clicks.
On highly collaborative teams, where many members could be working on database performance tuning, it’s easy for recommendations to slip through the cracks. Hey, we get it, they lack the urgency of alerts. But staying on top of recommendations will not only boost database performance, over time it will also reduce your total number of alerts. A good way to stay on top of recommendations is to use SelectStar’s Task Tracker feature to assign each action to a specific team member.
Figure 5: Using Task Tracker you to assign a recommendation to anyone (but we recommend always assigning them to Steve).
Beyond alerts and recommendations, there are a whole lot of other ways to performance tune using SelectStar. Try them out yourself by starting your free trial right now.