Growing to full-suite observability with Elasticsearch & the ELK Stack (PRT267)
summary
Dish Media describes moving from many point tools to a single Elastic stack pipeline that ingests 10 billion records a day from 25 million endpoints, uses Elastic Agent and Fleet for collection, and adds APM and OpenTelemetry for traces and metrics. The stack runs machine‑learning jobs for anomaly detection and forecasting, surfaces business KPIs alongside operational metrics, cuts investigation time from weeks to hours, and during a hurricane automatically detects signal loss and reroutes satellites to restore about 80% of affected customers.
AI-written summary · about
chapters (22)
- 0:00Intro
- 3:11The journey to full-suite observability
- 3:56We live in a world of exponential data growth and increasing complexity
- 4:52Advertising data lifecycle
- 7:26The everyday story before observability
- 9:44A solution that...
- 12:51Understanding operational and business KPIs
- 13:31Requiring monitoring across hybrid cloud
- 16:59Centralizing all data
- 18:36Ingest pipeline
- 20:11Elastic Agent: The single unified agent for logs, metrics, security, and more
- 21:34Collecting logs, events, and metrics
- 22:06Elastic APM: Flexible options for tracing
- 24:01Enrich the data for insights
- 25:52Benefits of full observability for insights
- 31:32Leverage comprehensive machine learning
- 31:56Go from reactive to proactive
- 35:45Machine learning applied to business data
- 37:29Observability and security: Start from development, not production
- 38:56The Dish dev cycle: Or do you test in prod?
- 41:31Security monitoring
- 45:56Before you go, here are tips to take with you
related talks
36mObservability from 0 to 100
19mScale observability and secure data in motion with Elastic and Confluent
37mBuilding Modern Observability Architecture With OpenSearch, OS Ingest...
40mDeploying OpenTelemetry Organizationally: From Proof of Concept to In-Production at Scale
34mMonitor Kubernetes cluster with Elastic
01h 00mElasticsearch: you know, for more than search
10mEffective Observability: Best Practices with Elastic
01h 02mDesigning an open-source observability stack for Kubernetes
