A survey of 100 VP+ IT decision-makers published today finds 62% plan to increase spending on observability initiatives over the next 12 to 24 months, with only 4% planning to decrease spending.
Conducted by LogicMonitor, a provider of a monitoring and observability platform, the survey also finds more than two-thirds (67%) are very likely or somewhat likely to switch observability platforms within one to two years.
A full 84% also said they are either actively consolidating tools (41%) or are considering it (43%). Nearly three quarters (74%) also said they are open to adopting a single observability platform if it meets their requirements, with 51% reporting they are already struggling with too many tools.
Not surprisingly, the survey also notes that 63% of respondents are also prioritizing artificial intelligence (AI) initiatives, followed by security (52%) and cloud infrastructure and services (36%). At the moment, however, only 12% of respondents said their organization is using AI for automating root cause analysis and remediation and only 4% said they have fully operationalized AI across IT operations.
More than half (52%), however, are investing in AI to accelerate root cause analysis and incident response, with nearly as many (47%) seeking predictive analytics to prevent incidents.
Karthik Sj, general manager for AI at LogicMonitor, said while there has in the last few years been a need for observability tools and platforms to enable DevOps teams to surface the root cause of an issue that has arisen in a complex IT environment, the rise of AI is about to more broadly force a deeper conversation.
Troubleshooting AI application environments requires access to massive amounts of telemetry data. On the plus side, generative AI technologies are making it simpler to interrogate that data via a natural language interface. However, AI applications and associated IT infrastructure also generate exponentially more telemetry data that many DevOps teams are already struggling to store and analyze.
It may be a while before AI is fully applied to AI, but in the short term major advances in reducing the amount of time required to investigate incidents are already being made, noted Sj. The goal now is to make it simpler for more IT teams to take advantage of the prompt and context engineering capabilities that are now being embedded into the next-generation observability platforms, he added.
The pace at which that transition will occur will naturally vary from one organization to the next but ultimately AI capabilities will soon become pervasively embedded across the entire software development lifecycle. Having those capabilities will then make it feasible for existing DevOps teams to manage the exponential amount of code that is already being generated using a wide range of AI coding tools, noted Sj.
Unfortunately, it may still be a while before DevOps teams acquire the tools and platforms infused with AI technologies that will be needed to effectively manage orders of magnitude more code that varies widely in terms of its quality. In the meantime, DevOps engineers will need to double down on their efforts to discover vulnerabilities and other weaknesses in code before it is deployed in a production environment, at least until more help in the form of more advanced AI capabilities finally arrives.
