The team that maintains the failover technologies has been building clustered storage for over 20 years so it is a seasoned team.  Expect the Scale Logic ZX to be class-leading among peers.  

  • SAS drives - dual ports on the storage devices themselves including flash devices
  • Dual Expander JBODs - dual expander backplanes allow SAS bus failover inside all chassis
  • Dual SAS Cables - MPIO enabled in the NAS allows 2 physical paths to each storage device
  • Dual nodes in HA mode - 2 identical NAS servers with one node acting as a hot spare for the storage pool operations
  • Cluster Resource Manager - uses STONITH ("Shoot The Other Node In The Head" or "Shoot The Offending Node In The Head"), a technique for fencing in computer clusters that prevents cluster split-brain and removes potential cluster instability
  • Network Bonds - bonded network interfaces allow ethernet cable failures
  • Network-based heartbeat and ping - used by the Cluster Resource Manager to ensure the health of nodes and promote or demote nodes by power cycling etc. 
  • Multi-Modifier protection - Base code developed by Lawrence Livermore Labs and funded by Intel - deep protection to prevent a pool from being imported by 2 nodes at the same time "split-brain" - https://openzfs.org/w/images/d/d9/05-MMP-openzfs-2017.4.pdf
  • Critical system error response policy - prevents cluster instability and triggers failover in case of unexpected hardware malfunctions.
  • Cluster watchdog - monitors volume availability and reboots the node to trigger pool failover in case of inaccessibility
  • IPMI hardware watchdog - monitors to guarantee a clean reboot or force a hung reboot for clean failover