Shared Responsibility Model For Platform Maintenance and Upgrades


Management of a HyperScale Reference Architecture platform is a shared responsibility. The customer is provided the choice of a preferred hardware platform which may require periodic updates as recommended by the OEM partner. HyperScale Reference Architecture software runs on several makes and models of servers where there are periodic updates to drivers and firmware. These updated firmware and driver releases are in turn certified on operating-systems for compatibility of the stack. Commvault, through its reference design program, validates a stack of components consisting of server hardware, firmware, drivers, OS and application software on the tested configuration. However, Commvault does not validate all combinations of the underlying components. A refresh of a previously validated reference design is performed only when there is a major design or component change. Changes that fall in this category include a new RAID controller card, new NIC/FC HBA requiring a new set of drivers, etc. If the customer chooses to upgrade a component of the HyperScale stack outside of the centralized Commvault upgrade process, the customer needs to take necessary steps to ensure compatibility and correct operation.

Note: HyperScale is dependent on specific device-drivers for correct operation. If there is a need to upgrade server firmware, the upgrade should exclude device-drivers.

Following is an illustration of this model of shared responsibility:



Hardware setup and installation


Firmware updates


Operating system and driver upgrades

Commvault releases the necessary updates periodically. Customer must check and install updates at regular intervals.

Commvault software updates

Commvault releases Feature Releases and Maintenance Releases at regular intervals. Customer must upgrade to the latest feature release and/or maintenance release on the HyperScale nodes.