Clusters for Machine Learning

page_header_divider_line

Overview

The potential of machine learning is pulling every organization in every industry into the realm of high-performance computing, many without the required systems skills and knowledge.  Bright Cluster Manager software automates building and managing machine learning clusters to reduce complexity, increase flexibility and accelerate time to value.  With a pre-tested catalog of popular machine learning frameworks and libraries, support for NVIDIA DGX Systems, GPUs and NGC containers, and integration with Jupyter Notebooks, Bright software ensures that data scientists, analysts and machine learning staff can have the computing environment they need for their demanding work.

product_page_header_image

Bright Cluster manager

Automated cluster
provisioning & setup

  • Provisions nodes from bare metal
  • Sets-up network, user directory & security
  • Configures and deploys Kubernetes
  • Includes pre-configured machine learning frameworks and libraries, Jupyter Notebook integration and supports NGC containers
  • Runs pre-checks across the entire cluster prior to deployment
  • Provisions nodes from bare metal
  • Sets-up network, user directory & security
  • Configures and deploys Kubernetes
  • Includes pre-configured machine learning frameworks and libraries, Jupyter Notebook integration and supports NGC containers
  • Runs pre-checks across the entire cluster prior to deployment

Automated problem
detection and isolation

  • Provides detailed monitoring metrics & health-checks
  • Reports resource usage by application and user
  • Allows automated corrective actions to be taken
  • Automatically detects and images replacement nodes

Automated change
management

  • 1-step propagation of updates
  • Auto-update nodes without reimaging
  • Easily add more servers & new components
  • Replace failing nodes without disruption
  • Prevents image-drift amongst nodes

Unparalleled
Flexibility

  • Supports mixed processor, accelerator and Linux combinations within the same cluster
  • Run jobs on bare-metal, VMs or containers in the same cluster
  • Extend to the public cloud for additional resources
  • Dynamically allocate resources where they’re needed based on demand
  • Re-image nodes on the fly to host new applications
page_section_divider_line
 
WATCH THE VIDEO: BRIGHT CLUSTER MANAGER AUTOMATES BUILDING & MANAGING LINUX CLUSTERS THAT SPAN FROM EDGE TO CORE TO CLOUD
Linux clusters are powering a whole new generation of business innovation. Powerful new hardware, applications and software have system administrators struggling with complexity, while providing  a system that is both flexible and reliable for end-users. Bright Cluster Manager eliminates complexity and enables flexibility by allowing users to deploy complete clusters over bare metal and manage them reliably from edge to core to cloud.
page_section_divider_line
WHAT'S NEW IN BRIGHT CLUSTER MANAGER 9.1

9.1 delivers more cluster management capabilities from edge to core to cloud, and comprises built-in expertise, integrated cluster build, management, and monitoring, and platform independence. Release 9.1 takes simplified cluster management to a whole new level. Here are some of the key features in 9.1:

  • Clusters-as-a-Service for VMware
  • Ansible Module
  • OpenShift Integration
  • Auto-scaling Enhancements
  • Jupyter Integration
  • Redfish Automated BIOS Management
  • Bright View UI Redesign
  • Offloadable Monitoring
 
Click the banner to see the latest updates to our enterprise-grade cluster management solution.
page_section_divider_line

Machine Learning Resources

VIDEO: BRIGHT CLUSTER MANAGER FOR MACHINE LEARNING

Learn how Bright Cluster Manager offers an integrated solution for building and managing ML clusters that reduces complexity, accelerates time to value, and provides enormous flexibility.

 

screenshot

REALIZE THE MASSIVE POTENTIAL OF YOUR AI INFRASTRUCTURE

Bright Cluster Manager on NVIDIA DGX Systems

Download Now

 

 

try it out