Table 5 MPX200 switch capability
Description
Switch capability
The 1 GbE iSCSI blade of the MPX200 comes with four copper GbE ports (802.3ab). To
take advantage of full duplex gigabit capabilities, you need infrastructure of Cat5e or Cat6
1 and 10 Gigabit
Ethernet support
cabling. The 10-GbE iSCSI/FCoE blade of the MPX200 comes with two SFP+ 10 Gb ports.
You can configure either SFP+ optical or SFP+ copper connectivity. Server connections and
switch interconnects can be done via SFP+ fiber cabling, in addition to Cat5e or Cat6
cabling, depending on IP switch capabilities.
For optimal switch performance, HP recommends that the switch have at least 512 KB of
buffer cache per port. Consult your switch manufacturer specifications for the total buffer
Fully subscribed
non-blocking
cache. For example, if the switch has 48 Gb ports. You should have at least 24 MB of
backplanesor Adequate
per-port buffer cache
buffer cache dedicated to those ports. If the switch aggregates cache among a group of
ports (that is, 1MB of cache per 8 ports) space your utilized ports appropriately to avoid
cache oversubscription.
IP storage networks are unique in the amount of sustained bandwidth that is required to
maintain adequate performance levels under heavy workloads. You should enable Gigabit
Flow Control support
Ethernet Flow Control (802.3x) technology on the switch to eliminate receive and/or transmit
buffer cache pressure. Note: Some switch manufacturers do not recommend configuring
Flow Control when using Jumbo Frames, or vice versa. Consult the switch manufacturer
documentation. HP recommends implementing Flow Control over Jumbo Frames for optimal
performance. Flow control is required when using the HP DSM and MPIO.
All ports on the switch, servers, and storage nodes should be configured to auto-negotiate
duplex and speed settings. Although most switches and NICs will auto negotiate the optimal
Individual port speed and
duplex setting
performance setting, if a single port on the IP storage network negotiates a suboptimal
(100 megabit or less and/or half-duplex) setting, the entire SAN performance can be
impacted negatively. Check each switch and NIC port to make sure that the auto-negotiation
is resolved to be 1000Mb/s or 10Gb/s with full-duplex.
It is important to enable Link Aggregation and/or Trunking support when building a high
performance fault-tolerant IP storage network. HP recommends implementing Link
Link
Aggregation/Trunking
support
Aggregation and/or Trunking technology when doing Switch to Switch Trunking, Server
NIC Load Balancing and Server NIC Link Aggregation (802.3ad).
Implementing a separate subnet or VLAN for the IP storage network is a best IP-SAN practice.
If implementing VLAN technology within the switch infrastructure, typically you need to
VLAN support
enable VLAN Tagging (802.1q) and/or VLAN Trunking (802.1q or InterSwitch Link [ISL]
from Cisco). Consult your switch manufacturer configuration guidelines when enabling
VLAN support.
In order to build a fault-tolerant IP storage network, you need to connect multiple switches
into a single Layer 2 (OSI Model) broadcast domain using multiple interconnects. In order
Spanning Tree/Rapid
Spanning Tree
to avoid Layer 2 loops, you must implement the Spanning Tree protocol (802.1D) or Rapid
Spanning Tree protocol (802.1w) in the switch infrastructure. Failing to do so can cause
numerous issues on the IP storage networks including performance degradation or even
traffic storms. HP recommends implementing Rapid Spanning Tree if the switch infrastructure
supports it for faster Spanning Tree convergence. If the switch is capable, consider disabling
spanning tree on the server switch ports so that they do not participate in the spanning tree
convergence protocol timings. Note: You should configure FCoE with spanning-tree disabled
at the first level server edge switch.
Sequential read and write, or streaming workloads can benefit from a maximum frame
size larger than 1514 bytes. The 1 GbE iSCSI and 10 GbE iSCSI/FCoE ports are capable
Jumbo Frames support
of frame sizes up to 9K bytes. Better performance is realized when the NICs and iSCSI
initiators are configured for 4K byte (maximum frame size of 4088 bytes) jumbo frames.
You must enable Jumbo frames on the switch, the 1 GbE iSCSI and 10 GbE iSCSI/FCoE
modules, and all servers connected to the IP-SAN. Typically, you enable Jumbo Frames
globally on the switch or per VLAN and on a per port basis on the server. Note: Some
switch manufacturers do not recommend configuring Jumbo Frames when using Flow
Control, or vice versa. Consult the switch manufacturer documentation. HP recommends
implementing Flow Control over Jumbo Frames for optimal performance.
24
Planning the MPX200 installation