This article discusses the benefits of a disaggregated observability (o11y) stack for modern distributed architectures, addressing issues of flexibility, high cost, and lack of autonomy in traditional solutions. It highlights key layers of a disaggregated stack — agents, collection, storage, and visualization — and suggests the use of systems like Apache Pinot and Grafana.
