An open-source toolkit for deploying and managing high performance clusters for HPC, AI, and data analytics workloads.
This patch release is focused on fixing following issue:
Installation of Kubernetes 1.16 and 1.19 are deprecated.
Spark Operator support is deprecated.
Omnia now installs Kubernetes 1.26
This release is focused on supporting following features:
Hardware Support: Intel E810 NIC, ConnectX-5/6 NICs.
Omnia github now hosts a “genesis” image with this functionality baked in for initial bootup.
Host aliasing for Scheduler and IPA authentication.
Login and Manager Node access from both public and private NIC.
Validation check enhancements:
Rearranged to occur as early as possible.
Isolate checks when running smaller playbooks.
Added a Benchmark Install Guide: OneAPI for Intel, MPI AOCC HPL for AMD.
This release is focused on supporting following features:
This release is focused on supporting following features:
This release is focused on supporting following features:
This release is focused on supporting following features:
Bugfix patch release which address the broken Singularity install issue.
Omnia has been enhanced to offer:
Omnia Prerequisites Installation
Provision Tool - xCAT installation
Node Discovery using Switch IP Address
Provisioning of remote nodes using
- Mapping File
- Auto discovery of nodes from Switch IP
Database Update of remote nodes info that includes,
Host or Admin IP
iDRAC IP
InfiniBand IP
Hostname
Inventory creation on Control Plane
iDRAC and InfiniBand IP Assignment on remote nodes (nodes in the cluster)
Installation and Configuration of:
NVIDIA Accelerator and CUDA Toolkit
AMD Accelerator and ROCm
OFED
LDAP Client
InfiniBand Switch Configuration with port split functionality
Ethernet Z-Series Switch Configuration with port split functionality
Bugfix patch release which address image retries and pod timeout.
Omnia has been enhanced to offer: