Why Today's APM Solutions Aren't Optimized for DevOps
July 21, 2014

Denis Goodwin
SmartBear

In my previous blog, Two APM Takeaways from Velocity Santa Clara 2014, I talked about fragmentation of solutions as a major trend in the APM space. Most vendors are providing a singular solution for some part of performance management, but yet, very few are able to provide a unified solution. In this post I want to extend this discussion to focus on what this means for DevOps.

I believe the lack of solution unification is an epidemic in the tech industry. Something that particularly stands out is that while more and more organizations are beginning to adopt the DevOps model by embracing continuous integration and collaboration, when it comes to the APM market serving these folks, some fundamental ideals of DevOps are thrown out the window — namely, the efficiency and collaboration aspects.

The Fragmentation of Tools is the Fragmentation of Teams

In the ever expanding vortex of value propositions and fragmented solutions that make up the APM market today, simplicity is getting lost. Everyone is so busy fighting to get or stay in the picture that they forget to take a step back to look at the big picture, and realize that all we are doing is collectively building an incomplete solution model. If it feels as though you are trying to integrate solutions that never were meant to integrate in the first place, it's because they weren't. And if the tools are disconnected, the teams that use them are much more likely to be disconnected as well. Or at least, not as connected as they could be.

When something breaks or a bug makes it into production, development and operations like to play hot potato with the blame — until everyone eventually tires and points their fingers at whichever one of their tools is most suspect for having failed them this time. However, the issue is often not the tool's fault either — the fault is shared by the assembly of tools that are collectively failing to support and encourage developers and operations to work together and collaborate.

Wait, Isn't DevOps About Efficiency?

It's natural to fix something after it breaks, but we are often not as good as we want to be at setting up prevention mechanisms to stop the issues from ever happening in the first place — perhaps if this weren't the case, APM vendors would have been building unified solutions from day one. This leads to businesses buying new tools for each part of a larger mission in order to alleviate some previous mishap and prevent it from happening again. The worst part is that by doing so you often can end up creating a much larger problem for yourself in the long term, as a result of opting for the quick fix in order to step around a small one in the short term.

Businesses are using an arsenal of tools that integrate (or don't integrate) with each other in attempt to assemble a complete APM solution. And then you have a whole other assorted toolbox for things like code review, functional testing, API testing and load testing for maintaining quality and superior user experience while continuously delivering applications. While these different solutions may be used by different functions, they are still all part of a single process. The more these silos are broken down in the future the more fluent and efficient that process will be.

Unified APM = DevOps

Perhaps the best approach is not about finding and filling the gaps with more and more tools as problems arise, and instead starting from square one with a single, unified platform for DevOps — one which fully supports collaboration between dev and ops, and enables ultimate efficiency by eliminating the need for added complexity and integrations.

The idea of eliminating the amount of time and energy spent evaluating, integrating and implementing all of these different tools makes the idea of unification worthwhile in itself, but of course this would only be the tip of the iceberg in terms of value.

In order to fully embrace the culture of DevOps and continuous integration, the tools that teams use need to support their fundamental principles and not dangle one leg over the fence. The days of patching a different piece of software onto every different part of the process are numbered.

Denis Goodwin is Director of Product Management, APM, AlertSite UXM, SmartBear Software.

The Latest

April 26, 2018

Moving more workloads to the cloud is a top IT priority, so eventually it will be time to consider how to make those critical legacy applications cloud ready. In Part 1 of this blog, I outlined the first four of eight steps to chart your cloud journey. In addition, consider the next four steps below ...

April 25, 2018

Clearly, moving applications to the cloud delivers significant advantages. So what's standing in the way of full cloud adoption? For many companies it's those burdensome (but critically important) legacy applications. Moving more workloads to the cloud is a top IT priority. So, eventually it will be time to consider how to make those critical legacy applications cloud ready. Consider the following eight steps to chart your cloud journey ...

April 24, 2018

Developers and engineering teams are under increasing pressure to release higher quality software faster. Continuous testing has proven to be central to these efforts as it helps eliminate bottlenecks and ensures that automated testing is a constant throughout the development process, not an exercise relegated to the "last mile." The value of automated testing is more evident than ever before, with nearly half the respondents reporting that management is fully committed to automated testing and with plans to increase spending, according to the recent Sauce Labs Testing Trends for 2018 report ...

April 19, 2018

As development speed has become a competitive advantage, the DevOps team has sought to enable continuous integration and continuous delivery (CI/CD). For the CI/CD process to be successful, it must be fast and efficient. Any potential roadblocks that delay any part of the process increase cycle times and slow down delivery ...

April 18, 2018

The top barriers to DevOps adoption involve stagnant organizational cultures; managing the jumble of legacy processes, IT infrastructure and newly created cloud environments; and growing software complexity that impacts application modernization initiatives ...

April 16, 2018

This is the third in a series of three blogs directed at recent EMA research on the digital war room. In this blog, we'll look at three areas that have emerged in a spotlight in and of themselves — as signs of changing times — let alone as they may impact digital war room decision making. They are the growing focus on development and agile/DevOps; the impacts of cloud; and the growing need for security and operations (SecOps) to team more effectively ...

April 12, 2018

Only 52 percent of developers using commercial or open source components in their applications update those components when a new security vulnerability is announced, according to new research conducted by Vanson Bourne for CA Veracode, part of CA Technologies. This highlights organizations' lack of security awareness and puts organizations at risk of a breach ...

April 10, 2018

For a few years now, it has seemed like agile developers and DevOps teams haven't been giving testing its proper due. One could almost picture them thinking, "So what if there's a bug, design flaw or performance issue. We'll fix it in the new version next week." Of course, this line of thinking has turned out to be a big mistake ...

April 09, 2018

Government IT professionals surveyed, as part of F5 Networks' State of Application Delivery report, made it clear they are heavily focused on building the foundation necessary for application-driven digital transformation. Around the world, government organizations are shifting towards digital government, and with that we see government organizations embracing the cloud, adopting automation and orchestration, and adjusting security strategies ...

April 05, 2018

The digital war room — physical, virtual or hybrid — is not in retreat but in fact is growing in scope to include greater participation from development and security. It's also becoming more proactive, with on average more than 30% of "major incidents" before they impacted business service performance. In this blog I'm providing a few additional highlights from the insights we got on digital war room organization and processes ...

Share this