Kubernetes Cross-Zone/Region Simplified: Harnessing Bacalhau for Efficient Distributed Compute
Talkat Kubernetes on Edge Day Europe 2025 · on Tuesday, 2025-04-01· at 11:10· Level 3 | ICC Capital Suite 10-12·
summary
Argues that moving compute to data cuts transfer costs and compliance risk, demonstrates an open source platform Bacalhau wired into the Kubernetes CRD so one command extends clusters to edge, running over a thousand simulated nodes (≈350 per region) and supporting containers, WᴀSМ/Wom and Java on devices from Raspberry Pi to 8‑GPU servers. Shows rolling thousands of ML jobs in place, updating models in situ, and reducing centralized ETL while preserving observability and auditability under GDPR.
AI-written summary · about
chapters (14)
- 0:00Introduction
- 1:03Infrastructure design flaws
- 4:56Cross-zone deployment necessity
- 5:59Compute over data architecture
- 7:26Efficient data processing scenarios
- 8:29Extending Kubernetes with Bakaya CRD
- 11:16Global job deployment demo
- 14:48Real-time job updates
- 16:32Distributed data warehouse use cases
- 18:40Near-compute data pipelines
- 19:44Data security and GDPR
- 21:31Technical requirements
- 21:54Summary
- 23:18Q&A
related talks
47mCloud Native and Sustainable, Reproducible Scientific Computing
07mData Gravity and Kubernetes: Managing Large-Scale Data Ingest With Minimal Latency
31mAn Adventure in Multi Cloud with Kubernetes and Distributed Data
23m08 Kubernetes and the cloud. Um, what's the cloud?
36mBuilding a Scalable Data Science & Machine Learning cloud using Kubernetes
08mContainerization – It’s Not Just for CPUs Anymore: Introducin...
08mLightning Talk: Orchestrating Kubernetes Clusters on HPC Infrastructure
24mDynamic Big Data in the Cloud with Kubernetes