DevOps: The Last 6 Months
2016 saw the popularity of DevOps explode across businesses of all sizes. And with adoption increasing from 66% to 74%, January saw an influx of articles predicting the challenges the developer community will face in the coming year.
Six months into the year, we reflect on some of those predictions and discuss where we are now:
More automation
According to Tech Republic: “[In 2016] companies began introducing tools to decrease the tedium of finding a line of buggy code in applications, and 2017 will see more automation for developers.”
We offer Maturity Assessments to our clients, during which we assess their existing DevOps and Continuous Delivery practices, to make company-specific recommendations for improvement. During this process, many companies see a lot of benefit in being able to automate manual tasks such as build, deployments, infrastructure provisioning and configuration.
Whether due to a lack of awareness (many of our enterprise customers don’t realise a number of their manual tasks could be automated) or that they’re sold on time they could save on business-as-usual and firefighting activities — automation continues to be an attractive offering to many businesses. And, as machine learning becomes more predominant, we’re likely to see the use of automation grow and grow.
Tool consolidation
Back in January, Joan Wrabetz, CTO at Quali commented: “The DevOps tools market is ripe for consolidation, mainly because of consumer demand and the sheer volume of companies out there.”
The market has indeed seen the effects of businesses looking to invest in only a small number of “best of breed tools”. We’ve seen our enterprise clients choose a few market leading tools to effectively enable legacy and slow delivery pipelines. It’s also evident in the number of acquisitions we’ve seen so far this year (think Cisco’s investment in AppDynamics and CA’s investment in Automic). Numerous other companies are also looking to increase their tool offerings to provide a broad range of capabilities to clients.
More containerization
Greg Willis, Director of Technology Operations and Systems Architecture at Morpheus Data, believes that in 2017: “We’ll see more enterprises start running their workloads on container platforms like AWS ECS, Kubernetes, DCOS, and Docker Datacenter”.
Over the past six months, our enterprise clients have been sitting up and discussing how to start transitioning to containers effectively. Whilst still a fairly new technology, benefits publicised by businesses making the most of containers are being heard by larger enterprises.
Although we’re a little way away from seeing containers become “the norm” in the enterprise, the efforts made by container solution providers (such as Docker who are beginning to address enterprise issues — think Docker Trusted Registry (security scanning of images), Notary (image signing) and UCP (enterprise orchestrator) make us think that widespread adoption is getting ever closer.
More security
Puppet showed in their 2017 State of DevOps Report that integrating security into software delivery positively effects continuous delivery. Many 2017 predictions focused on a “left shift” in security practices, that would enable businesses to continue increasing both speed and quality. Failure to do so would result in 2017 becoming the year of a major DevOps security breach.
Gartner reported back in 2015 that for 90% of companies using DevOps, security is an afterthought. But, we’ve seen that as practices have evolved and the speed of DevOps delivery has been proven, businesses are beginning to take more
consideration around security as part of the delivery lifecycle.
We encourage our clients to consider security at ground zero so they can embrace the philosophy of teamwork, coordination, agility and shared responsibility in their approach to security. We even wrote an article about it — which you can see here.
More monitoring (and more analytics)
Splunk CTO, Andi Mann, recently spoke at our All Hands on DevOps meetup, covering the topic of monitoring. As DevOps matures, he predicted, the process of getting performance feedback will receive more attention, driving better outcomes for DevOps teams.
Many of our customers are currently turning to applications such as Splunk and ELK Stack to measure and monitor their outputs. These applications prove invaluable across customers of all sizes as a way to measure success and prove that DevOps is both valuable to the business, and worth investing in further.
With a largely complex toolchain there are many moving parts. These could be tracked and analysed by individual tools however this causes a problem as all these tools do not provide collective data. An effective solution would be to aggregate your data to visualise and report on the entire lifecycle.
More cloud
David Linthicum, SVP at Cloud Technology Partners, suggested that 2017 will see a huge shift in businesses turning to the cloud: “If [businesses] moved 20 applications in 2016, then in 2017 they will move 500.”
Why? The myth of unsecure data in the public cloud is fast becoming a non-issue in the enterprise. Numerous banks including J.P Morgan Chase are now backing public cloud offerings — such as AWS and Microsoft Azure — to improve go-to-market speeds, strengthen security, improve customer experiences, and lower costs. The knock-on effect is that the perception of cloud risks is being cleared — meaning that other major businesses are taking on the cloud.
It’s great to see businesses embracing technology to realise new opportunities, and we look forward to supporting more enterprises as they take on cloud as a way to support DevOps.
An early trend we see with organisations moving to the cloud, is that many are not utilising the clouds full capabilities. We’ve seen organisations do a lift and shift approach rather than re-architecting to be cloud native. The cloud provides many benefits, beyond maintaining and configuring infrastructure, cloud providers offer functionality such as auto scaling, orchestration of infrastructure, native application hosting, geo-availability and so on.
More digital transformation
Last year marked the year that DevOps went mainstream. According to CA Technologies’ Aruna Ravichandran, 2017 will mark the emergence of DevOps at scale. 2016, saw mass adoption of DevOps practices, but this year businesses will drive DevOps across their entire software delivery lifecycle as part of their Digital Transformation efforts.
Following Cloud Expo Europe, we released an infographic to display the current state of DevOps and Digital Transformation in companies both large and small. It showed that this momentum is still continuing, with the DevOps adoption gap between SME and Enterprises narrowing and cultural change no longer a barrier to Digital Transformation.
Companies really are beginning to understand that in order to stay competitive, they have to get better at delivering software — and that’s where DevOps comes in.
In the coming years, we’ll continue to see the term DevOps overused and misunderstood. For us, whether you’re using the word DevOps or not, what’s important is ensuring you’ve made the necessary organisation culture changes to support your projects.
As DevOps becomes more and more popular, we’ll inevitably hear horror stories about companies that face blockers and ultimately fail at adopting culture. But that’s fine — it’s key for companies to fail to learn. Just make sure you’re failing fast — spot your failures faster — and learn from them just as quickly.
