Optimized primitives for collective multi-GPU communication

Kubernetes enhancements for Network Topology Aware Gang Scheduling & Autoscaling

Heterogeneous AI Computing Virtualization Middleware(Project under CNCF)

NVSentinel is a cross-platform fault remediation service designed to rapidly remediate runtime node-level issues in GPU-accelerated computing environments

NVIDIA Data Center GPU Manager (DCGM) is a project for gathering telemetry and measuring the health of NVIDIA GPUs

Kubernetes in Kubernetes

Fair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host or cloud, 400+ integrations.

Cloud operations platform

Kubernetes operator for multi-cluster metrics collection and analysis

Two conversational AI agents switching from English to sound-level protocol after confirming they are both AI agents

A tool to generate domain and project structures in openstack for testing purposes

Unified initial placement and scheduling of workloads within cloud and edge environments

Here lives the code for the energy aware cloud consolidation tool

cluster data collected from production clusters in Alibaba for cluster management research

Borg cluster traces from Google

Operations for the Ceph storage backend within the ApeiroRA stack.

Operations for the compute backend within the ApeiroRA stack.

Near-live migration toolkit for VMware to OpenStack

(Y)et (A)nother (K)ubernetes (E)ngine

KVM Node agent controlling Hypervisor objects to Kubernetes API

Lag-Llama: Towards Foundation Models for Probabilistic Time Series Forecasting

Kubernetes WithOut Kubelet - Simulates thousands of Nodes and Clusters.

Creates the REUSE metadata files (dep5, licenses) automatically depending on the declared licenses in an existing repository.

Cloud Native Infrastructure as a Service

Google Research