Observability and AI

Observability and AI are made to be partners. The integration of generative AI, particularly ChatGPT, into observability platforms has become a hot topic in systems engineering. Scott Moore, host of the SMC Journal podcast, explores this trend, analyzing early announcements from major observability vendors.

Let’s look at early adopters and their initiatives in 2023

New Relic: Pioneering AI Integration

New Relic took an early lead in March 2023 by announcing an “OpenAI quick start for New Relic.” While initially vague, this move signaled their commitment to leveraging AI technology. They followed up in May with the introduction of “Grok,” a generative AI assistant in limited preview.

Grok aims to democratize observability data access through a chatbot interface, allowing users to query information using natural language. This tool promises to simplify tasks like creating custom dashboards and understanding performance-related data points, potentially enabling users to generate simple dashboards with just two lines of code.

New Relic introduces Grok, a generative AI assistant:

https://siliconangle.com/2023/03/15/new-relic-brings-instant-observability-openais-gpt-3/

https://siliconangle.com/2023/05/02/new-relic-debuts-grok-generative-ai-assistant-observability-operations/

https://venturebeat.com/ai/new-relic-launches-grok-generative-ai-assistant/

Honeycomb: Simplifying Query Processes

In May 2023, Honeycomb debuted their “Query Assistant,” powered by OpenAI’s generative AI software. Unlike a chatbot, this tool focuses on simplifying the query process itself, helping developers understand and troubleshoot application errors more efficiently.

Honeycomb taps into chatGPT with Query Assistant: https://siliconangle.com/2023/05/03/honeycomb-introduces-generative-ai-query-tool-observability-platform/

https://devops.com/honeycomb-taps-chatgpt-to-simplify-observability/

Elastic: Powering Specialized AI Models

Elastic announced their “relevance engine” called ESRE in late May. This engine is designed to fuel generative AI models by facilitating the development of specialized models trained on structured and unstructured data. ESRE is expected to be integrated across Elastic’s product suite, including APM, enabling human-like queries for analyzing profiling and log data.

Elastic Elasticsearch Relevance Engine Will Fuel Generative AI Models

https://www.techdemand.io/news/tech-news/artificial-intelligence-news/elastics-elasticsearch-relevance-engine-will-fuel-generative-ai-models/

Dynatrace: Enhancing AI Automation

Dynatrace, known for its AI engine Davis, announced plans to enhance their platform with generative AI in late February. They’re focusing on “causal AI” to accelerate root cause analysis, addressing the challenge of consolidating massive amounts of data from multiple sources in complex environments.

Dynatrace Causal AI

https://www.forbes.com/sites/stevemcdowell/2023/02/19/dynatrace-blends-ai-automation—observability-with-new-offerings/?sh=6a5fb6757fe7

https://www.dynatrace.com/news/blog/productivity-innovation-with-chatgpt-generative-ai/

Datadog: A Different Approach

Interestingly, Datadog took a different route in May. Instead of launching a generative AI query tool, they introduced a new monitor to help organizations track their usage of generative AI models like ChatGPT. This monitor addresses concerns around cost, performance, error rates, and security.

Datadog

https://investors.datadoghq.com/news-releases/news-release-details/datadog-integrates-openai-chatgpt-help-organizations-monitor-ai

The Future of Observability and AI

While these early integrations primarily focus on simplifying data extraction through natural language interfaces, the true potential lies in a future where AI can proactively identify issues, create relevant metrics and dashboards, and even automatically resolve problems.

As the landscape continues to evolve rapidly, we can expect further developments and announcements from key players in the observability space. The race is on to leverage AI’s power in making observability data more accessible and actionable for a wider range of users.