A survey of 1,000 platform engineers and IT decision-makers in the U.S., United Kingdom and the Asia Pacific (APAC) finds that 22% claim to have been extremely successful, with another 72% reporting they have been moderate to very successful.
Conducted by Red Hat, the survey finds overall 59% of respondents have either established (41%) or advanced (18%) implementations of a methodology to centralize the management of IT services. A full 62% reported they have set up a dedicated platform engineering team.
The top two benefits cited by all respondents are increased innovation and increased developer satisfaction, tied at 35% each. Improved response times (34%), increased customer satisfaction (33%), increased deployment frequency (32%), improved security and compliance (32%) and increased developer productivity (31%), improved application performance (31%) and enhanced scalability and flexibility (31%) closely followed.
Surprisingly, the top reasons cited for adopting platform engineering are security vulnerabilities (48%), collaboration challenges (44%), toolchain complexity/gaps (42%) and compliance and regulatory requirements (41%).
Advanced automation (70%) is the top priority for survey respondents, followed by infrastructure configuration (53%) and monitoring/logs (53%).
The top investment areas are security and compliance (59%), developer productivity tools (55%), performance optimization (55%), cloud migration (54%), infrastructure modernization (53% and DevOps adoption (52%).
The top challenges being experienced are integration within existing workflows and security risks tied at 37%, followed by skills gaps (34% and budget constraints (33%), the survey finds.
The tip metrics being tracked are increased developer productivity (45%), improved security and compliance (41%), improved application performance (40%) and increased customer satisfaction (40%).
Finally, nearly half of respondents (45%) said generative artificial intelligence (AI) is a core element of their platform engineering strategy. Generative AI is already being used in platform engineering for generating documentation (76%), automating code generation (74%) and providing intelligent code suggestions (59%).
Mauricio Leal, a principal product manager for Red Hat, said the survey makes it clear that platform engineering is gaining traction whether everyone likes it or not. Many DevOps teams were initially created to provide application developers with more control over their environments. However, as the number of DevOps teams being created increased, organizations found they had deployed multiple types of the same tools, he noted.
IT teams are regularly now expected to do more with less, added Leal. No organization, for example, needs to deploy 18 variations of a continuous integration/continuous delivery (CI/CD) platform, he said.
Fundamentally, organizations are now discovering that rather than shifting control over everything left toward application developers in a way that inevitably overwhelms them, an application developer is a specific type of customer persona that can be enabled by a set of self-service capabilities provided by a centralized IT team. That level of self-service, however, requires increased investments in automation, said Leal.
Arguably, the biggest challenge when it comes to platform engineering will be changing the existing culture. Application developers are going to have to be convinced by example that a platform engineering team will best serve their interests. Application developers don’t necessarily want to assume responsibility for every task, but given past experiences with centralized IT teams many of them are, to say, the least going to be dubious of any effort that might threaten their prerogatives. The issue is finding a way to strike a balance between interests that are not as always aligned as they probably should be.