A question on the minds of many IT technologists is, “What will the data center look like in 2020?” Although that's only three years away, new technologies and services are rapidly changing the way application services are architected and delivered in the data center. Applications are impacted by Infrastructure as a Service, Software as a Service, Platform as a Service, serverless architectures, containers, and microservices. As such, these services will form the basis of the data center in 2020.
The challenge of managing tiered applications in this environment lies in the complexity that spans across not only the silos that exist within an IT organization, but across multiple service providers as well. It’s akin to troubleshooting across distributed systems that reside in their own ecosystems and are maintained by their own keepers. The people factor amplifies the magnitude of the problem, and grows exponentially, especially if there are IT seagulls around.
So what’s an IT seagull? An IT seagull is a person who swoops in on projects, takes a metaphorical “dump” on everything and everyone, and then flies off, leaving others to clean up the mess. Here's an example scenario: The IT team spends weeks on a hyperconverged infrastructure proof-of-concept for the internal data center. After much consideration and research, the team agrees on a vendor. But just before the purchase order is signed, an IT seagull flies in and says the team needs to go to the cloud because it's a "thing" now and executive leaders have bought into what the IT seagull just sold them. All of the team's due diligence and work goes out the window, but it still has to deliver everything according to the original timeline and budget.
Moreover, if a team's project is successful, IT seagulls will take credit for it all because they “touched” all parts of the project.
We all know an IT seagull; they tend to hover in management positions. Heck, we might even be guilty of being IT seagulls ourselves, given the pressures of being IT professionals who must remediate business-critical issues as quickly as humanly possible. Oftentimes, IT seagulls leave behind damage that eventually leads to the IT blame game. So how does one prevent IT seagulls from wreaking havoc on mission-critical projects?
There are three simple steps any IT professional in the midst of troubleshooting an application services issue can leverage to handle IT seagulls and minimize their damage.
- Stick to the data: Specifically, time-series data that can be correlated to root cause issues. Time-series data is a sequence of measurement points taken at successive, equally spaced time intervals. The key to troubleshooting is to quickly surface the single point of truth in order to eliminate the IT blame game. This effectively nullifies IT seagulls with data that correlates cause and effect in clear and concise terms.
- Collaborate: Good collaboration is another deterrent against IT seagulls. Being able to share a domain expert’s point of view with subject matter experts of other domains, and give them specific insights into a problem, is powerful when trying to quickly remediate issues across multiple stacks and service providers. It allows for decisive action to take place in the shortest amount of time and steps.
- Focusing on the connected context: By focusing on the connected context provided through correlated time series data that cuts across the layers of the entire application stack, teams can eliminate the IT seagull’s potential damage, even as they are busy dropping their lovely gifts from the sky onto our data centers.
The data centers of 2020 will be a hybrid of services provided by internal IT, as well as services sourced from third-party, best-of-breed providers such as Amazon Web Services and Microsoft Azure. These services will form the foundation for the application stack, and IT seagulls will flock to them like french fries on the beach. The application stack is central to the realization of any organization’s innovation, so it behooves enterprises be able to troubleshoot and remediate issues in the shortest amount of time possible—i.e., minimize mean-time-to-resolution (MTTR).
Accordingly, minimizing the effect of hybrid IT seagulls through correlated time series data and collaboration while monitoring with discipline will greatly optimize MTTR and application services delivery and consumption.