Deconstructing Apache Hadoop: A Dialogue About a Cloud-Native Refactor
summary
An Apache Hadoop refactor called Ozone is an S3‑compatible object store ported to Kubernetes (alpha v0.4, GA planned by Q3) that keeps HDFS semantics while exposing a CSI driver (FUSE now, NFS/native FUSE planned) and a control plane using gRPC and token-backed identities so only metadata servers need Kerberos keytabs. Prometheus metrics endpoints are added, tracing recommendation is OpenTelemetry/Jaeger, logs route via Fluentd/ELK, and a full service mesh is rejected as too many CRDs and too high latency overhead.
AI-written summary · about
chapters (5)
related talks
29mThe Future of Hazelcast
47mAll the DataOps, all the paradigms
06mIntegration and Ecosystem of Hadoop
27mWhy Building an Open Source IoT Cloud Platform Matters
25mAnkaios
32mA Fork to Reckon With: Minimizing Friction When Adopting OSS
26mAccelerating Innovation With Open Source - "Growing Banyan Trees"
25mMigrating a monolith to Cloud-Native and the stumbling blocks that you don’t know about.