DEVOPSdigest

DevOps experts — analysts and consultants, users and the top vendors — offer thoughtful, insightful, and sometimes controversial predictions on how DevOps and related technologies will evolve and impact business in 2018. Part 3 covers DevOps tools and teams.

Start with 2018 DevOps Predictions - Part 1

Start with 2018 DevOps Predictions - Part 2

UNIVERSAL DEVOPS PLATFORM

Backlash against the DevOps toolchain: Demanding a more unified approach, developers will speak out against the current DevOps toolchain. As opposed to logging into 10 or more different platforms to accomplish a task, developers will adopt a unified approach that is both seamless and effective. The handoff between multiple tools will ultimately dissipate with vendors offering a single product that runs all of the Dev and Ops in the same place.
Mark Pundsack
Head of Product, GitLab(link is external)

We'll see more companies with different solutions popping up, offering a wide menu of solutions for setting up agile infrastructure. The best solutions will combine the disconnected technologies into one consolidated enterprise platform.
Slava Koltovich
CEO, EastBanc Technologies(link is external)

VOLATILE DEVOPS TOOLS MARKET

Competition and complexity will increase as more vendors jump into the DevOps world. This will drive consolidation as vendors come and go. Enterprises will need to be careful about where they place their bets in leveraging cool new innovative technology that will disappear quickly.
Chris Boorman
CMO, Automic Software(link is external)

TRADITIONAL TOOLS ADD DEVOPS CAPABILITIES AND INTEGRATIONS

In 2018 we will see significantly increased adoption and evaluation of DevOps enabling tools by more enterprise IT organizations. This in turn will also put new pressure on the tool providers to deliver more features and capabilities to address common enterprise concerns around security, scalability, reliability, and role-based profiling. Traditional enterprise focused infrastructure and software providers will work to adopt more agility in their products to support growing demands for native integrations with DevOps tools. We will see a notable divide between providers who deliver enterprise features and integrations in a timely manner and those that do not.
Josh Atwell
Developer Advocate, NetApp(link is external)

ADAPTING DEVOPS TO LEGACY APPS

A major DevOps trend in 2018 will be the challenge of scaling early successful DevOps initiatives across the organization. In 2018, as more companies start to grapple with the complicated task of adapting DevOps to legacy applications, they will start embracing the strangler approach to breaking monoliths into manageable micro services. Such reinvestment will enable organizations to apply DevOps practices on the smaller components of legacy applications.
Ian Buchanan
Developer Advocate, Atlassian(link is external)

LEGACY APPLICATIONS VANISH

We'll see more organizations piloting containers, DevOps, and agile infrastructures for their greenfield projects, but only a minority will be from the migration of legacy applications to containers. Instead, legacy applications will vanish through rewrite and replacement of new apps over time. In tandem, there will be a lot of progress on data layer/storage management for containerized applications.
Slava Koltovich
CEO, EastBanc Technologies(link is external)

SHIFT LEFT PERFORMANCE

Performance and Quality aka Customer Experience were always important, but in today's digital native world they are exponentially more critical to your business. If a customer is trying to use or purchase your service and it takes too long or is too difficult to get to work, they are far more likely to lose patience and take their business elsewhere. By then, it's too late to recover and bad user experiences spread like wildfire in today's social media driven world. DevOps methodology, when fully embraced makes sure you can still deliver functionality quickly and be on top of usability issues before production deployment and lost customers. In 2018, shift left performance and quality will become more recognized as key attributes of a successful DevOps shop.
Scott Davis
EVP of Product Engineering & CTO, Embotics(link is external)

In 2018, we expect application performance (speed, availability) demands to reach even greater heights, which will further blur the dev and ops sides of the house where performance responsibilities are concerned. Increasingly, DevOps teams will institute strict performance requirements, starting earlier in the application lifecycle. If an application does not meet those benchmarks at a certain stage, it does not move forward — period. Once an application is in production, DevOps teams will require solutions that bring them together to identify the root cause of performance problems, resolutely and expeditiously. There's no more time for finger pointing.
Mehdi Daoudi
CEO and Founder, Catchpoint(link is external)

APPLICATION-CENTRISM

A modal shift in DevOps: from Infrastructure-centrism to Application-centrism. The growth in containerization, backed by the general shift to the cloud, and the increasing distribution and composition of applications means that there is a gradual shift from infrastructure-centrism, where the unit of value is a server, and the unit of work is a configuration, to application-centrism, where the unit of value is a service, and the unit of work is a deployment. This modal shift also requires automation at scale, though with different jobs to be done. DevOps tools need to match to these new modes, doing the new thing the right way, vs trying to force the old way. This change is constant, and the rise of serverless, or service-centrism will again adjust the mode and requirements for automation.
Marc Holmes
VP of Marketing, Chef(link is external)

BLENDING THE TEAM

We will see a blending of IT roles within Dev and Ops, creating more of an "engineer" position. For example, traditional roles such as Quality Assurance (QA), Sec, Ops and Database Administrators (DBAs) will blend together into engineering teams that are more product-focused.
Jason Hand
DevOps Evangelist, VictorOps(link is external)

SEPARATING THE TEAM

DevOps will bifurcate into a clear Dev role and a clear Ops role. This will be driven by the need for Dev to focus on innovating and coding. The Operational tasks of building environments, delivering masked test-data, building CI and enabling self-service to Dev will expand and merge with the IT Operations role.
Chris Boorman
CMO, Automic Software(link is external)

SHIFT LEFT HEADCOUNT

Op-headcount will shift left to Dev-headcount. As cloud infrastructure grows, Ops professionals will increasingly need to adopt software development lifecycle best practices to support IaC (Infrastructure as Code). Expect IT organizations to release headcount from their Ops departments and make new hires on Dev departments instead. These new developers will be hired as DevOps engineers, cloud engineers or CD pipeline engineers.
Scott Willson
Product Marketing Director, Automic Software(link is external)

BRIDGING THE GAP BETWEEN DEV AND OPS

As the need for optimized speed and bringing down Mean Time to Repair (MTTR) continues to grow, companies are looking for ways to become better at observability and getting a pulse on what's happening to their system without having to make a risky move. Given this, more companies will look to bridge the gap and meet in the middle between traditional ITSM and DevOps.
Jason Hand
DevOps Evangelist, VictorOps(link is external)

The need for speed will widen the chasm between Dev and Ops, threatening IT as we know it. As organizations strive for speed and efficiency in IT, developers are using a variety of tools and services to get the job done. Architectures like microservices are the backbone of new, dynamic apps, and containers are being used to speed deployment of the new functionality. This creates application complexity that makes today's static dashboards and human, manual assessments obsolete, forcing operations teams to find new tools and ways to ensure quality of service. Automation and intelligence are the answer. Automation provides discovery, insight and a model of everything that constructs the application, including logical and physical interdependencies between the microservices, middleware and containers. Artificial intelligence provides fast performance analysis, predicts potential incidents and offers remediation options. Automation and AI will help ops teams gain the speed to match that of modern application dev teams, bridging the chasm between them and advancing IT.
Pete Abrams
COO and Co-Founder, Instana(link is external)

Read 2018 DevOps Predictions - Part 4, covering Agile, CD and the development process.