Chapter 10. Copy services 629
The HyperSwap function uses a hyperswap topology by spreading the control enclosure of
the system across two sites, with storage at a third site that acts as a tie-breaking quorum
device:
The HyperSwap topology requires at least one control enclosure in each main data site.
Therefore, to get a volume that is resiliently stored on both sites, one control enclosure
with the necessary storage capacity is required.
The HyperSwap topology uses additional system resources to support a fully independent
cache on each site, providing full performance even if one site is lost.
The HyperSwap function can be configured by using the GUI or command-line interface
(CLI).
The hosts, Storwize V5000 control enclosures, and Storwize V5000 storage enclosures
are in one of two failure domains or sites. External virtualized storage capacity can also be
used.
Volumes are visible as a single object across both sites (the Storwize V5000 control
enclosure).
At least two control enclosures are required for HyperSwap. System scalability depends on
the hardware details, as shown in Table 10-12.
Table 10-12 HyperSwap support
So, a V5000 HyperSwap cluster is always restricted to a single control enclosure per site. The
Storwize V7000 and SVC can provide more scalability and offer more flexibility.
10.12.1 Introduction to HyperSwap volumes
The HyperSwap function is built on the Remote Copy features that include Metro Mirror,
Global Mirror, and Global Mirror with Change Volumes.
The HyperSwap function works with the standard multipathing drivers that are available on
various host types. No additional host support is required to access the highly available
volume. Where multipathing drivers support Asymmetric Logical Unit Access (ALUA), the
storage system informs the multipathing driver about the nodes that are in the same site and
the nodes that need to be used to minimize I/O latency.
The host and Storwize V5000 site attributes must be configured to enable this optimization
and to enable HyperSwap functionality. A three-site setup is required. Two sites are used as
the main data center to provide two independent data copies. A quorum disk or an IP-based
quorum can be used as a quorum device. However, the quorum device must be placed in a
third, independent site.
The quorum disk must be supported as an “extended quorum device”. The connection can be
implemented by using Fibre Channel, Fibre Channel through wavelength-division multiplexing
(WDM), synchronous digital hierarchy (SDH) and synchronous optical network (SONET), or
FCIP. The minimum bandwidth is 2 MBps.
V5010 V5020 V5030
/V5030
F
V7000 SVC
Maximum number of I/O groups 1 1 2 4 4
Support for HyperSwap No No Yes Yes Yes
Support for Stretched Cluster No No No No Yes
To Next Page
Loading...
Need help?
General