HP Integrity rx2800 HP Integrity Network Adapter Teaming Whitepaper - Page 51

General Description of Fast Path Operation

Page 51 highlights

The solution to this problem is the Fast Path mechanism designed and developed by HP Engineering. In addition to Fast Path detecting full connectivity loss, as described in the first example, Fast Path can also detect when a server's teamed ports have different hop count paths to the Core Switch (root switch). By listening to the path cost information contained in Spanning Tree BPDU frames reported by directly attached switches, a team can detect which switch is the best one to use for the team's Primary port. The result is that Fast Path provides the intelligence for a team to detect, and proactively react to, network events to minimize network latency for server data traffic. General Description of Fast Path Operation NOTE: This section assumes the reader has a technical understanding of the inner-workings of Spanning Tree - both IEEE 802.1D and Cisco's Per VLAN Spanning Tree Plus (PVST+). If additional information about Spanning Tree is required, refer to external sources (for example, "Interconnections: Bridges, Routers, Switches, and Internetworking Protocols" by Radia Perlman). Fast Path's operation can be simply described as the intelligence for a server to passively make teamed port usage decisions as if the server was a switch (without interacting with or affecting the behavior of Spanning Tree on the LAN in any way). This is accomplished by the teaming driver "passively" listening to Spanning Tree BPDUs being transmitted by the switches in the local network. BPDUs exchanged between switches enable them to effectively run the Spanning Tree algorithm to detect loops and bypass broken links within the network topology. Each of these Spanning Tree configuration BPDUs carry such information as the bridge ID (switch ID) of the designated root switch, the cost to reach the root switch, the bridge ID of the switch that transmitted the BPDU, the port number on the switch that transmitted the BPDU, etc. When a switch transmits these frames on a switch port that is connected to a server's teamed port and Fast Path has been enabled on the team, these BPDUs (and their information about the network topology) can be passively received and utilized by the teaming driver. In the example in Figure 4-10, an HP server with teamed ports is connected to two switches, Switch A and Switch B. The network administrator originally had both Switch A and Switch B connected directly to the Core Switch via Gigabit (1000 Mbps) links. This provided both Switch A and Switch B with equal access to the Core Switch and the rest of the network. As a backup, the network administrator directly connected Switch A and Switch B together via Link 3. Unfortunately, Switch A's uplink to the core switch was accidentally disconnected. This caused Switch A's effective bandwidth to the rest of the network to be reduced to 100 Mbps since all of Switch A's traffic must traverse Link 3 (100 Mbps). Based on the network configuration described in the previous paragraph, the teaming driver will receive two BPDUs, one from Switch A (on NIC 1) and one from Switch B (on NIC 2). The BPDU transmitted by Switch A to NIC 1 indicates a path cost to the core switch of 23 (cost of 19 for Link 3 plus a cost of 4 for Link 1), while the BPDU transmitted by Switch B to NIC 2 indicates a path cost to the core switch of only 4 (cost of 4 for Link 1). HP teaming will determine that Switch B provides the best path to the core network and will use NIC 2 as the team's Primary port. Types of HP Integrity Network Adapter Teams 51

  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15
  • 16
  • 17
  • 18
  • 19
  • 20
  • 21
  • 22
  • 23
  • 24
  • 25
  • 26
  • 27
  • 28
  • 29
  • 30
  • 31
  • 32
  • 33
  • 34
  • 35
  • 36
  • 37
  • 38
  • 39
  • 40
  • 41
  • 42
  • 43
  • 44
  • 45
  • 46
  • 47
  • 48
  • 49
  • 50
  • 51
  • 52
  • 53
  • 54
  • 55
  • 56
  • 57
  • 58
  • 59
  • 60
  • 61
  • 62
  • 63
  • 64
  • 65
  • 66
  • 67
  • 68
  • 69
  • 70
  • 71
  • 72
  • 73
  • 74
  • 75
  • 76
  • 77
  • 78
  • 79
  • 80
  • 81
  • 82
  • 83
  • 84
  • 85
  • 86
  • 87
  • 88
  • 89
  • 90
  • 91
  • 92
  • 93
  • 94
  • 95
  • 96
  • 97
  • 98
  • 99
  • 100
  • 101
  • 102
  • 103

The solution to this problem is the Fast Path mechanism designed and developed by HP
Engineering. In addition to Fast Path detecting full connectivity loss, as described in the first
example, Fast Path can also detect when a server’s teamed ports have different hop count paths
to the Core Switch (root switch). By listening to the path cost information contained in Spanning
Tree BPDU frames reported by directly attached switches, a team can detect which switch is the
best one to use for the team’s Primary port. The result is that Fast Path provides the intelligence
for a team to detect, and proactively react to, network events to minimize network latency for
server data traffic.
General Description of Fast Path Operation
NOTE:
This section assumes the reader has a technical understanding of the inner-workings
of Spanning Tree – both IEEE 802.1D and Cisco’s Per VLAN Spanning Tree Plus (PVST+). If
additional information about Spanning Tree is required, refer to external sources (for example,
“Interconnections: Bridges, Routers, Switches, and Internetworking Protocols” by Radia Perlman).
Fast Path’s operation can be simply described as the intelligence for a server to passively make
teamed port usage decisions as if the server was a switch (without interacting with or affecting
the behavior of Spanning Tree on the LAN in any way). This is accomplished by the teaming
driver “passively” listening to Spanning Tree BPDUs being transmitted by the switches in the
local network. BPDUs exchanged between switches enable them to effectively run the Spanning
Tree algorithm to detect loops and bypass broken links within the network topology. Each of
these Spanning Tree configuration BPDUs carry such information as the bridge ID (switch ID)
of the designated root switch, the cost to reach the root switch, the bridge ID of the switch that
transmitted the BPDU, the port number on the switch that transmitted the BPDU, etc. When a
switch transmits these frames on a switch port that is connected to a server’s teamed port and
Fast Path has been enabled on the team, these BPDUs (and their information about the network
topology) can be passively received and utilized by the teaming driver.
In the example in
Figure 4-10
, an HP server with teamed ports is connected to two switches,
Switch A and Switch B. The network administrator originally had both Switch A and Switch B
connected directly to the Core Switch via Gigabit (1000 Mbps) links. This provided both Switch
A and Switch B with equal access to the Core Switch and the rest of the network. As a backup,
the network administrator directly connected Switch A and Switch B together via Link 3.
Unfortunately, Switch A’s uplink to the core switch was accidentally disconnected. This caused
Switch A’s effective bandwidth to the rest of the network to be reduced to 100 Mbps since all of
Switch A’s traffic must traverse Link 3 (100 Mbps).
Based on the network configuration described in the previous paragraph, the teaming driver
will receive two BPDUs, one from Switch A (on NIC 1) and one from Switch B (on NIC 2). The
BPDU transmitted by Switch A to NIC 1 indicates a path cost to the core switch of 23 (cost of 19
for Link 3 plus a cost of 4 for Link 1), while the BPDU transmitted by Switch B to NIC 2 indicates
a path cost to the core switch of only 4 (cost of 4 for Link 1). HP teaming will determine that
Switch B provides the best path to the core network and will use NIC 2 as the team's Primary
port.
Types of HP Integrity Network Adapter Teams
51