At SC20: Bright Computing Launches HPC Cluster Manager 9.1

Print Friendly, PDF & Email

Today, Bright Computing, a specialist in Linux cluster automation and management software for HPC and machine learning, announced the latest version of Bright Cluster Manager (BCM) software. Version 9.1 is designed to simplify building and managing clusters from edge to core to cloud with the following features:

  • Integration with VMware vSphere allowing virtual HPC clusters to be created and managed without the need for skills and knowledge of vSphere
  • Integration with Ansible enabling administrators to build and manage Bright clusters using an Ansible playbook
  • Integration with RedHat OpenShift that uses Bright Cluster Manager in place of CoreOS
  • Deeper integration with the Jupyter platform, enabling Jupyter Notebooks to run on an entire cluster and support for HPC jobs running under workload managers
  • New auto-scaling enhancements that improve performance, provide more granular control, and improve ease of use
  • Integrated support for Redfish BIOS management
  • Ability to create dedicated monitoring nodes in a cluster that can support up to 100,000 compute nodes
  • A newly designed Bright View admin interface that improves navigation and usability

Bright Computing will be exhibiting this week at SC20’s Virtual Event. Stop by Bright’s virtual booth during exhibit hours for a demonstration on BCM 9.1 or additional information about our software. Along with our SC virtual booth, Bright has curated more content designed to help you build and manage your infrastructure from edge to core to cloud. Visit the SC20 Microsite and find additional speaking engagements, daily webinars, product launch demos, BCM 9.1 release demos, and partner activities.

“We are excited about the new features in 9.1,” said Martijn de Vries, CTO of Bright Computing. “Our development team has been hard at work implementing a host of features designed to better extend on-premise clusters to the public cloud and edge, improve ease of use, lower administrative costs, and increase standardization across the enterprise.”

With 9.1, we’ve expanded our market leading software for building HPC, machine learning, data science and hybrid cloud Linux clusters. Here are some of the key features in 9.1:

Cluster-as-a-Service for VMware – Bright Cluster Manager 9.1 provides an integration with vSphere that allows high-performance clusters to be created and managed without the need for skills and knowledge of vSphere.

Ansible Module – Bright is providing an integration with Ansible in the Bright Cluster Manager 9.1 release, allowing administrators to use their skills and knowledge of Ansible to build and manage Bright clusters using a familiar Ansible “playbook” approach.

RedHat OpenShift Integration – In Bright Cluster Manager 9.1 we are providing an integration with OpenShift that allows organizations to use Bright Cluster Manager to provide management of the underlying infrastructure in place of CoreOS.

Jupyter Integration – Bright Cluster Manager’s integration with Jupyter accomplishes several important things that make Jupyter a more effective and powerful tool for users.

  • Bright makes it possible for Jupyter Notebooks to run on a cluster
  • Bright’s integration with Jupyter provides a point-and-click interface for users that are not familiar with the complexity of submitting jobs to a cluster.
  • Includes support for HPC workload managers
  • Users also have control over the environment that their notebook is running in, such as where and how their kernels are run, the number of tasks, the consumable resources, the job name prefix, the directory the kernel runs in, and which queue the job is submitted to.

Auto-scaling Enhancements – Bright Cluster Manager 9.1 includes enhancements to auto-scaling that improve performance, provide more granular control and improved ease of use.

Redfish Automated BIOS Management – Bright Cluster Manager 9.1 is introducing support for Redfish BIOS management. This allows Bright administrators to view and update BIOS parameters on any/all nodes of the cluster that support the Redfish specification, from the command line or from the BrightView GUI.

Offloadable Monitoring – In Bright Cluster Manager 9.1, the cluster monitoring function for the entire system can be “offloaded” from the system’s head node to a set of dedicated servers that perform system monitoring exclusively, freeing the head node to perform its other duties and allowing the system to continue scaling to 100,000 nodes.

Bright View UI Redesign – The Bright View graphical admin interface has been redesigned to improve navigation and usability.

Source: Bright Computing