High Availability HPSS

High Availability (HA) HPSS is a configuration of the standard HPSS offering that adds redundancy and failover capability to ensure continued data access in the event of hardware, software, or site failures. It includes enhancements to both Core Servers and Movers, working together as part of a unified HA solution.

See HPSS Offerings for offering details and Contact Us with any questions you might have about the use of High Availability HPSS.

High Availability Core Servers

The Core Server portion of HA HPSS includes the configuration of a standby HPSS Core Server and associated Db2 metadata storage, as well as the configuration and testing of HPSS Core Server failover-with-manual-approval functionality.

Db2 log shipping protects HPSS metadata against a broad range of failures including hardware issues, network disruptions, and even full-site outages. It helps minimize downtime in the event of such incidents. Db2 log shipping replicates the HPSS Db2 metadata to a standby HPSS Core Server, which can be installed locally or even at a remote facility.

HA Core Server details include:

  • Two HPSS Core Server machines: one primary and one standby
  • Metadata on the primary server is synchronized to standby server
  • Negligible impact on performance
  • Primary computer failure results in the standby computer assuming the IP/Hostname of the service interface and restarting HPSS

High Availability Movers

High Availability Movers provide access to HPSS storage when an HPSS Mover server fails. HA Movers are typically deployed for the HPSS disk cache only, as HA Movers for tape drives are not considered cost-effective at scale. HA Movers are always deployed in pairs, so the pair’s operational Mover can be used to manage access to the storage devices for the failed Mover.

HA Mover services include the configuration and testing of the HPSS Core Server failover software for each Mover.

HA Mover details include:

  • Shared access to the Mover storage for each pair of Mover servers.
  • Primary computer failure results in the pair’s operational Mover computer managing access to the storage devices for the failed Mover.