Understanding and monitoring the performance and health of applications and systems is critical. This is where observability comes into play.

Observability is about gaining a comprehensive understanding of a system's internal state by analyzing the data it generates within the context of real-time events and actions across the infrastructure. Unlike traditional monitoring, which might only tell you when something has gone wrong, observability equips you with the tools and insights needed to discern not just the occurrence of an issue but its underlying causes. This enables teams to proactively manage and optimize their systems rather than merely reacting to failures. 

From another perspective, as the central hub for managing and routing traffic across distributed services, the API gateway provides a comprehensive view of how applications are consumed across an enterprise. By implementing observability at the API gateway layer, organizations can ensure better performance, security, and visibility across their entire API ecosystem. Here are some key benefits:

• API Performance

  • Comprehensive Monitoring: Observability enables tracking of request patterns, response times, and error rates across all applications.
  • Faster Issue Detection: Quickly identify high latencies, unexpected error rates, or traffic spikes before they impact downstream services.

• API Security

  • Enhanced Security & Troubleshooting: Acts as a first line of defense by monitoring and logging suspicious activity, failed authentication attempts, and potential attacks.

• API Consumption

  • Distributed Tracing: The API gateway can initiate and maintain distributed tracing by injecting trace IDs into all requests, enabling seamless tracking across services.
  • Service Dependency Insights: Since API gateways route requests to multiple applications, observability helps map dependencies, providing insights into failures and bottlenecks.

• Business & Performance Metrics — Gain valuable insights, such as:

  • Most accessed APIs
  • Usage patterns by different clients or regions
  • API performance trends over time
  • Effective monitoring of SLAs and compliance metrics

For AI and LLM-based services the Gateway provides metrics per provider, model, embeddings, etc. allowing cost calculation as well as control over AI tokens counted by LLM providers.

To achieve this, observability relies on three primary pillars:

• Logs provide detailed, timestamped records of events and activities within a system, offering a granular view of operations.
• Metrics deliver quantitative data points that reflect various aspects of system performance, such as request count, error count, response times, and throughput.
• Traces provide high fidelity data on sampled requests. They provide a detailed overview of exactly which components a request traversed, allowing teams to visualize and analyze the end-to-end journey of transactions and interactions.

With this context in mind, let's explore how Kong Konnect and Dynatrace work together to build advanced observability processes using OpenTelemetry (OTel) standards.

OpenTelemetry

OpenTelemetry is an open source observability framework designed for cloud native software, formed from the merger of the OpenTracing and OpenCensus projects in 2019. Since then, OpenTelemetry has evolved to be the open standard approach to instrumenting, generating, collecting, and exporting telemetry data such as metrics, logs, and traces.

OpenTelemetry signals come from many sources such as traces and logs, auto-instrumented code libraries or custom instrumentation as well as metrics from 3rd-party vendor platforms. These signals can then be sent to external observability solutions for in-depth analysis and visualization.

While OpenTelemetry represents an amazing set of standards for how to ingest, modify and egress instrumentation data, it does not actually provide a platform for storing, visualizing and analyzing this data.

Dynatrace 

Dynatrace simplifies cloud complexity and enables faster and more secure innovation. The Dynatrace platform combines broad and deep observability, monitoring, continuous runtime application security, and advanced AIOps with Davis® AI to provide answers and intelligent automation from data. This enables innovators to modernize and automate cloud operations, deliver software faster and more securely, and ensure flawless digital experiences.

Dynatrace fully integrates with these open standards, such as OpenTelemetry. As a leading committer to OpenTelemetry, Dynatrace works closely with the open source community to drive ongoing improvements, using our decades of experience in observability for insight into customer challenges and best practices.

Kong Konnect

Kong Gateway is a fundamental component of the Kong Konnect API management infrastructure, known for its platform-agnostic approach, scalability, and outstanding performance, all driven by its plugin-based extensibility. It centralizes critical functions such as proxying, routing, load balancing, and health checking, effectively managing both microservices and traditional API traffic.

A key strength of Kong Gateway is its extensibility, supported by a wide range of plugins that enable the implementation of various policies, including Authentication/Authorization, Rate Limiting, Proxy Caching, Requests and Responses Transformation, and Traffic Control.

Kong Konnect and Dynatrace integration

For observability, Kong provides the OpenTelemetry plugin to propagate traces, Logging plugins to externalize transactional logs and the Prometheus plugin to produce metrics to any OpenTelemetry compatible server. Below is a reference to the architectural layers of an application using Kong and how it might be observed with a combination of OpenTelemetry and Dynatrace.