I own a Jeep these days and I love it. One thing about Jeep owners is that we tend to be the type that don’t mind doing maintenance and repairs ourselves. Having been around garages since I could walk, I’m no stranger to dirt and grime on my hands as a result of simple tasks like oil changes. Bigger jobs, including diagnosing complex problems, though, often require special equipment and knowledge that, frankly, I don’t possess. So, that’s when I take the Jeep to a mechanic.
When it comes to vehicles, we all have the choice of doing needed maintenance and repairs ourselves or hiring someone to do it for us. But few people bother doing their own car repair or maintenance, opting instead to have someone do it for them. It’s easier, you know it will be done correctly, and you can arrange your schedule around the work being performed.
These same principles apply to data and databases. Much like how a vehicle is vital to most people’s everyday life, databases contain the most important asset your company owns: data. Without data, your company probably wouldn’t exist. All that hardware you administer -- the servers, routers, switches, cables -- exists because of the data. Therefore, database performance is foundational to your company’s operations.
Furthermore, like Jeep owners, most businesses usually take a do-it-yourself attitude towards their data. As such, many companies, including yours, may have a database administrator (DBA) on staff, maybe you’re even just such a database professional. If so, you know that data professionals are expected to be able to solve just about any problem, with the tools they currently have.
But with today’s feverish pace of technological advancement, it is becoming harder to stay on top of the latest data technologies. That’s why ideas such as DevOps are becoming more and more popular, because it’s no longer sufficient to work in silos, where the DBA, the network administrator, the systems administrator, and a host of other IT roles work largely independently. Today, constant communication across teams is a must.
Data professionals also must be able to diagnose issues quickly and present their findings to a variety of teams, speaking in terms that each team can understand. In the new hybrid IT world, data professionals must especially be able to identify if an issue they’re experiencing is related to one of the many networks their organization may be reliant upon, whether owned or that of a service provider.
In fact, network latency is the resource that matters most to the cloud DBA because network latency will determine how to build solutions that meet the recovery point objective and recovery time objective metrics set by the business. As hard as it may be to pinpoint an issue with my Jeep, it can be hard to find that one poorly written query consuming so much network bandwidth that the rest of the business slows to a crawl.
I previously covered several tips on just how to accomplish this, but my point here is that as hybrid IT becomes mainstream, we are going to find that breadth is the new depth. Data, network, and system administrators are going to need to know a little about a lot. DBAs are not going to have all the tools and knowledge necessary to fix or tune every possible thing that could go wrong. Working together across teams with a single point of truth surfaced by comprehensive monitoring and management has to become the new normal.
In the end, all that will ever matter is that the right data is in the hands of the right people at the right time. That’s what will make all the difference in the world for you and your infrastructure teams.