PVST/PVST+
Generating a tree per VLAN is a straightforward and simplest solution. It guarantees that there is no loop in every VLAN. However, for various reasons, the spanning tree protocol working in this way does not form a standard, but each vendor has a set, especially represented by Cisco's VLAN spanning tree PVST (Per VLAN Spanning Tree).
In order to carry more information, the format of the PVSTBPDU is different from the STP/RSTPBPDU format, and the destination address of the transmission is changed to the Cisco reserved address 01-00-0C-CC-CC-CD. Moreover, in the case of a VLAN trunk, the PVST BPDU is tagged with an 802.1Q VLAN. Therefore, the PVST protocol is not compatible with the STP/RSTP protocol.
Cisco quickly introduced the improved PVST+ protocol and became the default spanning tree protocol for switch products. The improved PVST+ protocol runs on VLAN 1 as the normal STP protocol, running the PVST protocol on other VLANs. The PVST+ protocol can communicate with STP/RSTP. The spanning tree state on VLAN 1 is calculated according to the STP protocol. On other VLANs, the common switch only forwards the PVST BPDUs as multicast packets according to the VLAN number. However, this does not affect the elimination of the loop. It is only possible that the root bridge status of VLAN 1 and other VLANs may be inconsistent.
Figure 5 PVST+ and SST docking diagram
The default VLAN of all links in Figure 5 is VLAN 1, and both VLAN 10 and VLAN 20 are trunked. SW1 and SW3 run the single spanning tree SST protocol, while SW2 runs the PVST+ protocol. On VLAN 1, it is possible that SW1 is the root bridge and port 1 of SW2 is blocked. On VLAN 10 and VLAN 20, SW2 can only see its own PVSTBPDU, so on these two VLANs it considers itself to be the root bridge. The PVST BPDUs of VLAN 10 and VLAN 20 are forwarded by SW1 and SW3. Therefore, when SW2 detects this loop, it will block VLAN 10 and VLAN 20 on port 2. This is the STP/RSTP compatibility provided by the PVST+ protocol. It can be seen that the Layer 2 loops in the network can be identified and eliminated, and there is no point in forcing the consistency of the root bridge.
Since each VLAN has an independent spanning tree, the defects of the single spanning tree are overcome. At the same time, PVST brings new benefits, that is, Layer 2 load balancing.
Figure 6 PVST+ load balancing diagram
In Figure 6, all four devices run the PVST+ protocol and both have VLAN 10 and VLAN 20. Assume that SW1 is the root bridge of all VLANs. It can be configured to block VLAN 10 on port 4 and VLAN 20 on port 2. The link of port 1 of SW4 can still carry the traffic of VLAN 20. The link of port 2 can also carry the traffic of VLAN 10 and the function of link backup. This is not possible in the case of the previous single spanning tree.
The smart PVST/PVST+ protocol implements VLAN cognition and load balancing capabilities, but new technologies have also brought new problems. The PVST/PVST+ protocol also has their “unspeakable”.
The first flaw: Since each VLAN needs to generate a tree, the traffic of the PVST BPDU will be proportional to the number of VLANs of the trunk.
The second point is that when the number of VLANs is large, the amount of calculation and resource consumption for maintaining multiple spanning trees will increase dramatically. In particular, when the state of the interface of many VLANs changes, the state of all spanning trees must be recalculated, and the CPU will be overwhelmed. Therefore, Cisco switches limit the number of VLANs used, and it is not recommended to trunk many VLANs on a single port.
Third point defect: Due to the private nature of the protocol, PVST/PVST+ cannot be widely supported like STP/RSTP. Devices from different manufacturers cannot directly communicate with each other in this mode and can only be implemented in some flexible ways. For example, Foundry's IronSpan. IronSpan runs STP by default. When a port receives a PVSTBPDU, the spanning tree mode of the port is automatically switched to PVST/PVST+ compatibility mode.
In general, the topology of the network does not change frequently, so these shortcomings of PVST/PVST+ are not very fatal. However, the need for a large number of VLANs for port trunks still exists. As a result, Cisco has made new improvements to PVST/PVST+ and introduced a multi-instantiated MISTP protocol.
Multi-instantiated third-generation spanning tree protocol: MISTP/MSTP
The Multi-Instance Spanning Tree Protocol (MISTP) defines the concept of "Instance". Simply put, STP/RSTP is port-based, PVST/PVST+ is VLAN-based, and MISTP is instance-based. The so-called instance is a collection of multiple VLANs. The method of bundling multiple VLANs into one instance can save communication overhead and resource occupancy.
When used, multiple VLANs of the same topology can be mapped to an instance. The forwarding state of these VLANs on the port will depend on the state of the corresponding instance in MISTP. It is worth noting that the VLAN and instance mapping relationships of all switches in the network must be the same, otherwise network connectivity will be affected. In order to detect such an error, the MISTBP PDU carries the information such as the VLAN relationship of the instance in addition to the instance number. The MISTP protocol does not process STP/RSTP/PVST BPDUs, so it is not compatible with the STP/RSTP protocol. It is not even compatible with the PVST/PVST+ protocol. Loops occur when the network is deployed together. In order to allow the network to smoothly migrate from PVST+ mode to MISTP mode, Cisco has also implemented a hybrid mode MISTP-PVST+ that can handle PVST BPDUs in the switch product. When upgrading the network, you need to set the device to MISTP-PVST+ mode, and then set it to MISTP mode.
The benefits of MISTP are obvious. It has both PVST VLAN awareness and load balancing capabilities, as well as low CPU usage comparable to SST. However, the poor backward compatibility and the privacy of the protocol block the wide range of applications of MISTP.
The Multiple Spanning Tree Protocol (MSTP) is a new multi-instance spanning tree protocol defined in IEEE 802.1s. This agreement is still in the process of continuous optimization, and now only the draft version is available. However, Cisco has added MSTP support in CatOS 7.1. Huawei's Layer 3 switch product, the Quidway series, is also coming soon with a new version that supports the MSTP protocol.
The MSTP protocol is exquisite in that MSTP-enabled switches and non-MSTP-switched switches are divided into different areas, called MST regions and SST domains. Run the multi-instance spanning tree inside the MST region and run the RSTP-compatible internal spanning tree IST (Internal Spanning Tree) on the edge of the MST region.
Figure 7 Schematic diagram of MSTP working principle
The switches in the MST region in the middle of Figure 7 use MSTP BPDUs to exchange topology information. The switches in the SST domain use STP/RSTP/PVST+ BPDUs to exchange topology information. On the edge between the MST region and the SST region, the SST device considers that the docked device is also an RSTP device. The state of the MST device on the edge port will depend on the state of the internal spanning tree, which means that the spanning tree status of all VLANs on the port will be the same.
The spanning tree that needs to be maintained inside the MSTP device includes several internal spanning tree ISTs, and the number is related to how many SST domains are connected. In addition, there are a number of MSTP spanning trees determined by multiple spanning tree instances (MSTI). The number of instances is determined by how many instances are configured.
Compared with the previous spanning tree protocols, MSTP has obvious advantages. MSTP has VLAN-awareness and load balancing. It can implement fast RSTP-based port state switching. You can bundle multiple VLANs into an instance to reduce resource usage. The most valuable thing is that MSTP is well backward compatible with the STP/RSTP protocol. Moreover, MSTP is an IEEE standard protocol and the resistance to promotion is small.
It can be seen that the omnipotent MSTP protocol can be the consistent direction of today's spanning tree development is well deserved.
The future of the spanning tree protocol
The development of any technology will not be stalled by the emergence of an "ideal" technology, and the development of the spanning tree protocol itself illustrates this. As the application deepens, various new Layer 2 tunnel technologies emerge, such as Cisco's 802.1Q Tunneling, Huawei's Quid way S8016's QinQ, and MPLS-based Layer 2 VPN technology. In this new situation, users and service providers will have new requirements for spanning tree protocols. Where should the Spanning Tree Protocol go? Although there is still no unified answer to this question, vendors have begun active exploration in this area. Perhaps in the near future, the spanning tree protocol supporting Layer 2 tunneling technology will become the standard protocol for switches. You may also need information about ccie v5 syllabus, click hypelink to read more.