Does it replace Bright Cluster Manager?

Yes. A migration guide is provided. The difference is environment-specific optimization plus Suite integration.

Luxe Series · L4

CLOSED PoC

Luxe Orbit

A specialized tool for OS & SW provisioning and parallel management of heterogeneous compute resources. It delivers systematic resource management for tens to hundreds of compute nodes through simultaneous deployment and parallel processing.

Request a Closed PoC Get the technical brief

01PAIN POINTS

Solving these challenges

The operational hurdles teams hit most often in the field — see how Luxe Orbit resolves them in the LIVE DEMO.

01

A single management-server failure affects operation of the entire cluster

02

Redeploying the OS across ~200 nodes takes days

03

Failure logs are scattered per node, making unified tracing difficult

02FEATURES

Feature details

Key capabilities organized by category. Each group is delivered as a working scenario in the LIVE DEMO.

F.01

Management-Server HA

Active-Active redundancy fails over within 30 seconds, preventing cluster downtime from a single failure

Active-Active cluster configuration (based on Pacemaker + Corosync)
Real-time synchronous replication of the management DB, image catalog, and node status
30-second automatic failover with split-brain prevention
Asynchronous replication to a DR site (optional)

F.02

Node Boot Automation

Boot 200+ nodes simultaneously within 12 minutes using PXE, iPXE, kickstart, and cloud-init

Full support for PXE, iPXE, and HTTP boot (compatible with both UEFI and Legacy)
kickstart (RHEL, Rocky, Ubuntu) and cloud-init support
OS images, packages, and user hooks managed together in a single manifest
Batch remote power control via IPMI and Redfish
Simultaneous BMC firmware update mode — choose rolling or blackout

F.03

Image Catalog & Node-Group Policies

Golden-image version control and rollback guarantee environment consistency across the whole cluster

Version-controlled image catalog (Rocky, Ubuntu, RHEL, SUSE)
Per-node-group environment variables, boot hooks, and post-install scripts
Rollback workflow — automatic revert to the previous image
Lifecycle management — scale up, scale down, reconfigure, decommission

F.04

Unified Failure Logs

Search syslog, dmesg, IPMI SEL, and journald logs from a single unified index

Unified collection of per-node syslog, dmesg, and journald via Fluent Bit
IPMI SEL and Redfish event logs indexed alongside
Chronological unified view — trace cascading failures across nodes
Choice of OpenSearch or Loki backend (to match operational policy)

03SCENARIO

Flagship scenario · integrations

Flagship scenario

Boot 200 nodes simultaneously

12 min

A full-cluster OS redeploy in 12 minutes. Management-server failover in 30 seconds.

INTEGRATIONS

Systems you can integrate

Slurm
Spectrum Scale
NVIDIA Mellanox UFM
Luxe Ray

04FAQ

Frequently asked questions

Q. Does it replace Bright Cluster Manager?
Yes. A migration guide is provided. The difference is environment-specific optimization plus Suite integration.

Next steps

Need infrastructure design consulting to make this product possible?

Request a Closed PoC View related Solutions