HP 3PAR StoreServ Concepts Guide: HP 3PAR OS 3.1.3

Overlapping QoS Rules
A VV can be a member of multiple VVsets, each of which can have a QoS rule defined. In such
a case, the I/O to and from volumes in the VVset is governed by multiple, possibly overlapping
rules. All active rules for a particular VVset are combined using wired-OR logic: the QoS limit that
is reached first takes precedence. A QoS rule can be created on a VVset that has none, or not all,
of its VVs exported to a host.
Minimum QoS Settings
HP 3PAR Priority Optimization provides no enforcement for minimum QoS levels. The minimum
will be ensured, provided the system has been correctly sized for the expected total IOPS and
throughput, and provided QoS rules against all VVsets were installed such that their combined
sum does not exceed what the system is capable of delivering.
QoS on Copied Volumes
Virtual and physical copies of VVs in a VVset are not automatically a member of the parent VVset;
they have to be added manually to the parent VVset in order to be governed by the same QoS
rules. Virtual and physical copies can be subject to a different QoS rule than their parent VV, if
desired. QoS rules are persistent across a reboot of a HP 3PAR StoreServ Storage system. A VVset
cannot be removed unless all QoS rules defined for it are removed first.
Automatic Transparent Failover
The purpose of an automatic transparent failover is to migrate the service of host I/O from the
failed primary to the surviving secondary storage system. In an automatic transparent failover,
there can be no coordination of operation between the storage systems. Only those actions
performed on the secondary system are executed. The operation initiation sequence also differs
somewhat from a manual transparent switchover:
• In a manual transparent switchover, the operation is initiated by issuing the setrcopy
switchover command, on a remote-copy volume group basis, on the primary storage system.
• In an automatic transparent failover, the trigger is provided to the surviving secondary storage
system by the Quorum Witness infrastructure and related software, which begin state transition
and activation of the secondary host paths for all applicable secondary volume groups.
The outcome of both the manual and the automatic peer-persistence operations will be the same,
in that the original secondary system will begin to service host I/O to the failed-over volume groups;
but in an automatic failover, replication will have stopped, and will remain unavailable until the
failed system is recovered.
Automatic Transparent Failover 63