The Right Toolchain Is the One You Build

One of several iterations that IT likes to go through is corporate standards for technology versus best tool for the job. We like to have standardization because it allows staffing portability between projects, but we like right tool for the job because from the moment you choose a standard, competitive factors and changes in needs make the standard less of a big win.

DevOps is still struggling through this process, particularly in enterprises where standardization bears a bit more weight on important projects.

The fact is that DevOps is well suited to the idea of iterating standards–somewhere between a wild west of everyone using their “right tool” and an organization dictating the tool of choice.

Most of us use Jenkins, because it does what we need it to do. What needs doing is different from organization to organization–some use it for continuous testing, while others go all the way through application release automation. But the fact is we could iterate and replace some, or all, of the functionality we’ve decided to use Jenkins for tomorrow. Kind of like a “We have a standard that is evaluated regularly” approach.

For those new to DevOps, the array of choices is downright bewildering. Xebia Labs has a toolchain diagram page to help those already doing DevOps, and several sources have spreadsheets/PowerPoint templates to help you work through your existing environment. While all of the options these tools try to organize for you are great for the industry to find better solutions, it makes for a confusing mess. Your toolchain could be all AWS with some scripting, all Jenkins with plug-ins or 15 different apps working independently. A surprising number of organizations are even using Git for the bulk of their toolchain.

What matters is that you know what you have, and know what your options are. Build the best toolchain for your environment today, and keep half an eye on it to be aware when something more suited to your environment comes along. We still maintain python scripts that do a fair amount of our automation, but several toolsets in place also.

There are objectively better tools out there, but for our purposes they are almost universally overkill. So we go with what’s best for our environment. As time has gone on, we’ve integrated more and scripted less, but it will be a while before out toolchain is actually just tools–if ever.

If your toolset is doing the build/test/deliver/deploy cycle well enough, you’re covered. If it’s not, you need to look for alternate solutions. Even if it is, make sure you’re aware of what’s on the market and not just what’s “hot” right now. Be aware of those companies that are up and coming, maybe in spaces that are already flooded. What differentiates them, and in our environment would that differentiation matter? That’s all you need to know–there is no need for a super deep technical evaluation, unless you’re actually considering them. Just enough to know if their differences add value.

Keep cranking out the apps that make business run. Perhaps with an updated toolchain, perhaps not. In the end, what works for your environment is all that matters.

Don Macvittie

Don Macvittie

20 year veteran leading a new technology consulting firm focused on the dev side of DevOps, Cloud, Security, and Application Development.

Recent Posts

Who’s Responsible for Security? Apparently, It Depends

More than 10 years after organizations began implementing DevOps, responsibility for security still resembles the proverbial chicken and egg dilemma.…

5 hours ago

How Embedded Real-Time Engagement Makes the Internet More Inclusive

We hear the word “inclusivity” thrown about quite a bit these days. Some would call it a buzzword, but inclusivity…

6 hours ago

BMC Software Closes Compuware Deal, Sets Course

BMC Software today announced it has formally completed its acquisition of Compuware as part of an expanded effort to modernize…

18 hours ago

PagerDuty Looks to Streamline IT Incident Management

PagerDuty has added more automation and analytics capabilities into its portfolio of incident management tools, at a time when organizations…

18 hours ago

BMC Completes Acquisition of Compuware to Mainstream Mainframe DevOps

Customers to benefit from an integrated DevOps toolchain with unparalleled mainframe operations management and agile application development and delivery HOUSTON, June…

20 hours ago

5 Reasons Why Your App Needs an Error Monitoring System

All software applications have errors. Bugs are simply part of software development. That’s why engineering organizations need to know exactly…

1 day ago