Kubernetes Batch + HPC Day Europe 2022
2022-05-17
Valencia, Spain
[Kubernetes Batch + HPC Day Europe 2022, Kubernetes Batch + HPC Day Europe 2023]
as the organisers put it
register explore the schedule VIEW ATTENDEE GUIDE That’s a wrap on Kubernetes Batch + HPC Day at KubeCon + CloudNativeCon EU 2022! Thank you to all the attendees, sponsors and speakers that joined us…
who it's for
Kubernetes platform engineers, cloud operators, HPC engineers, machine learning engineers, data engineers, cluster operators, research computing managers
topics
GKE custom autoscaler; GPU slicing and orchestration; Kueue Queue and ClusterQueue; NodeResourceTopology NUMA support; topology-aware scheduler plugin; Volcano batch system features; CRI Resource Manager for CPU pinning; Apache Pulsar tiered storage pipelines; Ludwig AutoML with Ray autoscaling; YuniKorn scheduler plugin and hierarchies
| title | author | length |
|---|---|---|
| Opening + Welcome | Abdullah Gharaibeh, Ricardo Rocha | 06m |
Half-day program running until 5 pm with a Google Cloud keynote, five 25-minute sessions and three lightning talks, focused on making Kubernetes a better… | ||
| Keynote: High Performance Computing on Google Kubernetes Engine | Maciek Różacki | 07m |
Google claims GKE runs production HPC at scale, supporting clusters up to ~45,000 nodes and shipping a custom autoscaler that optimizes for customer cost. The… | ||
| Kueue: A Kubernetes-native Job Queueing | Abdullah Gharaibeh | 35m |
Kueue is a Kubernetes-native queuing controller that actually enforces job-level admission using the existing batch/v1 Job API: it introduces Queue (namespace… | ||
| Resource Orchestration of HPC on Kubernetes: Where We Are Now and the Journey Ahead! | Swati Sehgal, Francesco Romani | 26m |
A NUMA-aware system is implemented: a NodeResourceTopology CRD plus a node agent (NFD topology updater or custom resource topology exporter) and a… | ||
| Volcano – Cloud Native Batch System for AI, BigData and HPC | William (LeiBo) Wang | 24m |
Cloud-native batch system that adds job-level lifecycle, multi-pod templates and queue decoupled from namespaces to Kubernetes, with concrete integrations for… | ||
| Get More Computing Power by Helping the OS Scheduler | Antti Kervinen, Alexander Kanevskiy | 20m |
Adds CRI Resource Manager (CRI‑RM) to nodes to take over CPU and memory pinning via the CRI host manager, replacing direct kubelet→container runtime calls and… | ||
| Fast Data on-Ramp with Apache Pulsar on K8 | Timothy J Spann | 14m |
Shows an actual NiFi -> Pulsar pipeline that ingests REST and database sources with no code, pushes them into Pulsar topics, runs lightweight Pulsar functions… | ||
| How to Handle Fair Scheduling in a Private Academic K8s infrastructure | Lukáš Hejtmánek, Dalibor Klusacek | 07m |
CERIT-SC runs a single multi-tenant Kubernetes cluster with 2,500 CPU cores, 130 users, ~1,000 pods/day alongside an HPC system of 30,000 CPU cores, 3,000… | ||
| Efficient Deep Learning Training with Ludwig AutoML, Ray, and Nodeless Kubernetes | Anne Marie Holler, Travis Addair | 28m |
Runs Ludwig AutoML and Ray on Kubernetes with Nodeless K8s (Luna) to provision just‑in‑time GPUs, replacing a fixed three‑node g4dn.4xlarge Ray cluster… | ||
| Apache YuniKorn A Kubernetes Scheduler Plugin for Batch Workloads | Wilfred Spiegelenburg, Craig Condit | 30m |
Apache YuniKorn graduates to an Apache top-level project and offers a new plugin execution mode that runs YuniKorn as scheduler plugins to the default… | ||
| Closing | Aldo Culquicondor | 03m |
Conference closes with an update: the kubernetes working group is new but meets every other week with active contributors and maintains a Slack channel and… | ||
| nothing matches | ||