Showing results for 
Search instead for 
Did you mean: 

Seven Ways to Love Your Data

Level 17


I spend a lot of my time with data. Too much time, really. An almost unhealthy amount of time.

I read about data. I write about data. I tweet about data.

I'm all about the data, no trouble.

And because I spend so much time immersed in data I see a lot of mistakes out there. A lot of issues that are avoidable. Data breaches caused by a lack of adequate data security. Disasters made worse by not having backups. Data centers offline but nobody being alerted because the alerting is not configured correctly.

Avoiding these issues is simple: Hire someone with a working knowledge of data, databases, and data administration. Of course, good help is hard to find. In the absence of finding good help, you need some guidance about what you can be doing better.

So that’s what I want to do for you today. Here are seven things that you can do to love your data, starting today. Not all the items listed here can be done in a day. Some will take longer than others. Review the list, build a plan, and get started on making your data the best it can be.

Automation – As a data professional you need to spend time thinking of ways to automate yourself out of a job. I would even suggest that you must be thinking about ways to “cloud-proof” your job. GUIs, and people, don’t scale. Code does.

Security – Data is the most critical asset your company owns. Without data, your company would not exist. All that hardware your company owns? Yeah, that’s there to move bits of data back and forth. When you consider the value and criticality of your data, you will understand that it is necessary to deploy data security and privacy tools.

Maintenance – In an unscientific study I did last year, I found that the bulk of database performance problems was related to the lack of proper maintenance. In some cases, there is no maintenance being done at all, or it is the wrong maintenance (such as VM snapshots instead of database dumps).

Alerting – Many folks lump alerting together with monitoring, but they are different. You use monitoring to decide what you want to alert upon. Alerts require action; everything else is just information you can collect and use later.

Monitoring – Here’s the most important thing you need to know about monitoring: If you create inbox rules for your monitoring system, then you’ve already lost. Monitor and measure what you need, but only send emails when action is needed (see above).

Analytics – While we are talking about alerting and monitoring, here’s your reminder to learn some basic data analysis and use it against the monitoring data. Learn to spot outliers. Stop guessing about what happened and start learning about what will happen.

Backups – It’s 2017. Do you know where your database backups are? But it’s not just backups. Last month the cloud went down. Be prepared. Have a proper business continuity plan in place and test the process twice a year.

Successful data professionals use these seven items to exercise proper control over their enterprise. Further, many of the issues I read about all fall into one of these seven buckets.

The difference between being prepared and being unprepared comes down to your willingness to make sure you have each of these seven items covered.

Level 16

wow, another perfect article amazing write up
Level 16

awesome sqlrockstar

Level 19

So true that many of these perils are avoidable.  I think we're moving closer to getting it right these days.  Maintenance of the datacenter is extremely important too... are the batteries still good in the HUGE UPS... are the belts on the blower for cooling unit moving air across the water from the chilller still good?

Level 18

Very good write-up..

Having the backups is one thing...being able to spin up a local DB as a business continuity plan is another. 

Are the backups where you can get to them easily and in a timely manner ?

Level 21


Level 17

Great article. The biggest problem with backups is that most don't test them. So when they really need them, they don't work.

Level 13


Level 10

Hmm...while dependency on cloud services is in the rise, I wonder how we will address issues like latency, HA, alerts and monitoring in the near future.

All in all, a very good article.

Level 15

Nice article. Seems you left the best for last. Backups are soooo very important.

Level 16

Where does Retention fall into these categories? Given this month's Thwack mission data retention is an extremely important issue. Compliance, folks...

Level 10

If you create inbox rules for your monitoring system, then you’ve already lost....

I have been fighting this issue for years, and trying to get my hands on the fools flooding my inbox. I no longer feel so alone.

Level 21

Great write up sqlrockstar​; I especially liked the bits about Monitoring and Alerting.  I can't tell how frustrating it is to hear somebody to ask for something to be monitored and assume that they need alerts also.  When I ask them why they need alerts they always tell me "well, I just want to be aware of it".  Do people not realize they can actually login to the monitoring system and look at it?

Level 13

I like your comment about the inbox rules related to monitoring.  I have too many rules set up for items that I don't care about.  Some of them from back in the day when I worked in different departments and was never removed from the distribution group.

Level 13

I'm not sure I can love my data...

maybe Data...but not Lore

Level 17

Great question! I've been trying to raise awareness for about a year now that the future for a Cloud DBA is to be able to understand and troubleshoot network issues first before trying to tune a query. After that, the Cloud DBA will need to help architect the right HA solution for the business based upon cost/benefit/risk, and then monitor to make certain that RPO and RTO can be met at all times, which also means being able to monitor the network.

Level 17

I've just resigned myself to the fact that we are data hoarders. We rarely have a proper archiving strategy. We want all the data, at all times.

Level 18

so much data, so little time, so few cycles to process...

About the Author
Thomas LaRock is a Head Geek at SolarWinds and a Microsoft® Certified Master, SQL Server® MVP, VMware® vExpert, and a Microsoft Certified Trainer. He has over 20 years experience in the IT industry in roles including programmer, developer, analyst, and database administrator.