As we start a new year, we look toward the future and the emerging trends that are set to shape the world of software delivery for 2017 and beyond.
One of the most powerful, overarching trends we saw in the recent year is the growing adoption and maturity of DevOps practices. From its grassroots, small-teams beginnings, DevOps has mushroomed and is now at the point of becoming the standard way of working (or rather, the pursuit and continued improvement towards this way of working) – for enterprises and startups alike.
[ What are the signs of great vs. mediocre DevOps organizations? See our related article, DevOps Jobs: How to spot a great DevOps shop. ]
For large organizations, we see that what has started as a “keeping up with the unicorns” grassroots movement within more forward-thinking companies, has matured to large, complex enterprises now often being on the forefront of DevOps innovation. These enterprises are now leading the way in terms of patterns and best practices for scaling DevOps across the entire organization. The question is — why?
Fueled by competition, adaptability is key
To borrow a lesson from Charles Darwin, it is not the strongest or most intelligent that survives — it’s the ones that are most adaptable to change.
Today’s highly competitive and dynamic business landscape gives rise to new business models and offerings – all driven by software innovation.
Software is disrupting and creating new value streams across all industries – new and old. The example we’re all familiar with is Amazon, which radically transformed the retail sector and was massively influential in ushering in an age of an e-commerce revolution. Retail businesses today must respond to Amazon and the changed expectation and experience in the industry, and have an effective online retailing solution. Apple’s disruption of the music industry, Uber’s disruption of the taxi industry and countless other examples, are all a testament to how software is changing every aspect of our lives, across all industries.
Today’s market shows that simply standing pat on your prior accomplishments is a poor strategy for future success. This is also demonstrated in the S&P 500 Index: If you go all the way back to 1958, the average tenure for a company on this prestigious list was about 61 years. Today, the average tenure is only 18 years — and projections indicate that period will be as short as 10 years by the year 2030.
To compete, lead, and stay in the lead in today’s digital era, organizations must invest in software innovation, and release application updates at a much faster pace. To become better at developing and releasing software – they MUST invest in DevOps.
Adapting, downtime, and the bottom line
As you are no doubt keenly aware of, downtime costs real dollars. But how much, exactly, does downtime cost your business? Recently, IDC Research released a report that, for the average Fortune 1,000 company, downtime costs organizations $100,000 an hour (for an infrastructure failure). For a mission-critical application, that figure skyrockets to $500,000 to $1,000,000 per hour.
In today’s disruptive world, minimizing downtown is a key consideration to being competitive. Undergoing a DevOps transformation holds a lot of benefits where it comes to enable businesses to ensure uptime and increase release cadence and quality.
The 2016 State of the DevOps Report shows that high-performing IT organizations employing DevOps practices are deploying 200 times more frequently than their low-performing counterparts. Correspondingly, these organizations are also able to enjoy 24X faster recovery from failure and minimize downtime.
Microservices and containers will be driving modern application delivery
With all these organizations making a change toward DevOps, microservices are a uniquely aligned architecture to help achieve success. This is because microservices enable organizations to architect their solutions around a set of decoupled services — that can each be developed and released independently. Each service focuses on doing just one thing well and enabling more rapid time to market, with less interdependencies to reduce risk.
Going hand-in-hand with microservices, containers serve as an ideal deployment vehicle. That’s because containers are designed to run one isolated process, with minimal deployment and runtime overhead. What do the stats say about this? Again, Gartner expects that by the end of 2018, 50 percent of enterprises will be running containers in Production.
Everything to the left
Across the board, companies are emphasizing shifting-left of more and more of their software delivery practices. As part of this trend, which will continue to increase, activities that were traditionally done after deployment and production, or things that are typically done later in the development or release process, are now moving earlier in the pipeline (or, to the left).
The goal of this left-shift is to identify issues earlier, thus reducing their impact and cost. This means things like automated and continuous testing – including performance testing and disaster-recovery testing/practice – are being done earlier in the cycle.
Big data and DevOps coming together
The next trend to look for in 2017 is big data and DevOps coming together to create predictive analytics throughout your delivery cycle. One massive thing DevOps tools have accomplished is automation — automating the process as well as automating the configuration. In these pipelines, you’re suddenly creating a ton of data. When you start applying machine learning to this, you can really start to hone in on some interesting data where you can predict failures and identify areas for optimization in your pipeline.
Consolidation becomes critical for scale, compliance, utilization
As DevOps matures, large, complex organizations will focus on finding ways to consolidate and standardize their DevOps processes, tooling and implementation – in order to scale DevOps throughout the organization (and save costs).
To enable developers and an agile way of working, while ensuring governance, system-level visibility, and organizational control – many are implementing a shared, self-service DevOps Automation platform to enable their end-to-end pipelines. Rather than investing in each team’s software delivery as a “snowflake” set-up, they invest in converging of all teams and applications around shared configurations, pipelines, environments, tooling, processes, security test, etc. This consolidation allows for reusability, improved visibility, security, auditability, and resource utilization – while still being flexible enough to support specific teams’ needs.
In this fragmented, dynamic and evolving market, organizations need to be ready to support a continued “hybrid” complex state of existence: for their infrastructure, architectures, tooling, processes, release pipelines, and so on.
This is particularly true for large, complex organizations, who will need to continue to support monolithic/legacy applications alongside microservices; on premise alongside cloud infrastructure; VMs and containers; traditional application releases alongside CD-type pipelines, etc. This continued state of “hybrid-everything” poses challenges on large-scale operations, and will put continued emphasis on DevOps solutions and patterns that are “future-ready” to support any tool, technology stack, process, or application in this hyper-hybrid state.
Finally, culture is king
Finally, one of the biggest trends for 2017 will not be just a focus on engaging and implementing some of these DevOps best practices into your enterprise, but a sweeping adoption of the DevOps/agile culture. This is because one of the most important – if not the absolute most key –tenets to a successful DevOps organization is culture. The enterprises that most espouse the shared responsibility, the empowered autonomous teams, the can-do attitudes, and the continuous learning environment in which DevOps thrives will see the biggest benefits.