In today’s age, users demand instant response times for any application they use.

You may have the best product on the market, but if you can’t load the app within milliseconds to any user around the world, you will never be a unicorn!

After years of investigating this, it turns out that performance problems are either the database or the network.

So let’s see how to rule out the database!

Data Performance problems

Performance problems are never just as simple as “It’s the database” or “It’s the network”.

The Data Guardian will be called in to look at any slow-down to the database.

This is because Data Performance is one of the three key technical areas for the Data Guardian.

And without the tools in place to visualize performance over time, it will be difficult to pinpoint the root cause with any certainty.

And without that historical context, it is difficult to put any long-term plans in motion to prevent future issues.

Data Paramedics

If you have ever had to call in an emergency health issue, it is likely that paramedics showed up to triage the situation.

They come in and try to stabilize you so you can be moved to the hospital.

As part of that, they will take various measurements of “vitals”.

Vitals are the key signals to the health of the body. Things like blood pressure, pulse, and dilated eyes.

Unfortunately, these vitals are taken as a one-time snapshot. There is no history of what is normal for the person they are stabilizing.

And so, they will ask questions of anybody with that person to help fill in gaps of what is occurring.

This can mean it takes longer to gather the necessary information to stabilize the patient.

The person in charge of the data environment who has no historical visibility of the performance of the database is like those paramedics.

They must quickly assess the situation by looking at the vitals of the database to triage performance issues.

Database vitals are the CPU, IO and memory usage. There are also database statistics to review to try to understand what is happening.

They try to piece together the context around the performance issue.

And while going from emergency to emergency, these data paramedics do not have time for planning for long-term performance solutions.

Data Doctor

Contrast that with a general physician, who you go to at least once a year for a checkup.

This doctor has access to your medical records, and they take new measurements of your vitals. They can also do more extensive tests to get even more data of your health.

Recommendations of better eating and regular exercise serve as proactive strategies to keep the performance of your body high.

The Data Guardian should work to be more like the doctor to have systems in place to collect regular measurements across server and database vitals.

Of course, collecting database vitals once a year would never be enough. So it must be done more frequently.

These historical measurements provide the context of the environment and can help you make a better diagnosis faster.

This context can also allow the Data Doctor to make proactive recommendations to prevent performance issues before they are issues.

Conclusion

As the person in charge of your company’s data, you will likely be called in to stabilize many performance issues.

Ideally you will have familiarity of the vitals of a well functioning data environment.

You should set yourself up for success by having historical visibility into that system.

So in this way, the Data Guardian should be a Data Doctor instead of a Data Paramedic.

This will allow you to plan ahead and keep your users from the pain of suffering data performance problems.