Filter
Top Products
• Low round-trip latency—In order to not impact disk I/O to the application server, the round-trip
latency between the sites must be no more than 2 ms.
• Redundant links—Have multiple physical connections (media) between the sites for redundancy.
The network should be configured so that a link failure does not cause the Multi-Site SAN to
go offline.
• Assign servers to sites—Avoid high-latency connections by assigning application servers to
the site where the server is located. Application servers assigned to a site connect only to
storage systems in the same site as follows:
◦ Without MPIO—With servers set up in the CMC with the load balancing option, the
system sets up the iSCSI session only to storage systems assigned to the same site.
◦ With MPIO and the HP StoreVirtual DSM for Microsoft MPIO—The system sets up iSCSI
sessions only to storage systems assigned to the same site.
◦ With MPIO and the Microsoft DSM—With servers set up in the CMC with the load
balancing option, the system sets up the iSCSI session only to storage systems assigned
to the same site.
• Use the Failover Manager—Unless you have storage systems in three physical sites, use a
Failover Manager. Put the Failover Manager in the third site (physical or logical) located on
a network that is isolated from the other sites and has a different power connection. Minimum
requirement: 100 MBit/s with 5 MS latency. For information about the Failover Manager,
see “Using the Failover Manager” in the HP StoreVirtual Storage User Guide.
Common Multi-Site SAN network designs
Based on the Multi-Site SAN configuration, there are many network designs that can be implemented
that solve the redundancy, high availability, and performance requirements for the Multi-Site SAN.
A few commonly used designs are:
• Dual redundant links between the sites
• Full-mesh triangular (three) redundant links between sites
• Hub and spoke central network core with redundant links out to each site
• Combination full-mesh core network with triangular (three) links between sites
Balancing the requirements for high availability during certain types of failures in the Multi-Site
SAN and the budget associated with the network infrastructure typically dictates the network
architecture you choose to implement. The most redundant network typically will cost the most to
implement. However, not implementing redundancy in the Multi-Site SAN network significantly
curtails the benefits of the Multi-Site SAN software. Table 1 (page 6) shows the common Multi-Site
network configurations and the types of failures against which they provide protection.
Table 1 Common Multi-Site network configurations and failure protection of each
Storage system failure
protection
Network failure protectionSite failure protectionNetwork topology
YesYesYes, with manual
intervention
Dual redundant links
between sites
YesYesYesFull-mesh triangular links
between sites
YesYes, so long as the core
does not fail
YesHub and spoke central core
YesYesYesCombo full-mesh core with
triangular links between sites
6 Designing a Multi-Site SAN