PCE+

Background

The ingress runs the constrained shortest path first (CSPF) algorithm and uses information stored in the traffic engineering database (TEDB) to calculate MPLS TE tunnels. On an inter-domain network, each ingress can only obtain topology information within a single domain. Therefore, the ingress faces the following challenges when establishing inter-domain tunnels:

PCE+ solution can help resolved the preceding issues in MPLS networks. This solution involves two device roles:

• PCE server: usually an SDN controller. A PCE server stores the path information of the entire network and computes paths based on stored information to optimize network-wide resource usage.

• PCE client: usually an SDN forwarder serving as a tunnel ingress. A PCE client is the initiator of path computation requests. After receiving the path computation results and tunnel constraints from a PCE server, a PCE client sets up a TE tunnel as required.

Benefits

Related Concepts

PCE server

Defined in relevant standards, a PCE server is an entity that can use network topology information to calculate paths or constrained routes. A PCE server can be an operations support system (OSS) application, a network node, or a server. A PCE server on an MPLS TE network receives a calculation request sent by an ingress and uses TEDB information to calculate an optimal constrained path for an MPLS TE tunnel.

PCC

A path computation client (PCC) sends a calculation request to a PCE. The ingress of an MPLS TE tunnel can function as a PCC.

PCEP

The Path Computation Element Communication Protocol (PCEP), defined in relevant standards, exchanges information between a PCC and a selected PCE and between PCEs in different domains.

Domain

A domain can be an Interior Gateway Protocol (IGP) area or a Border Gateway Protocol (BGP) autonomous system (AS). The NE20E supports IGP areas only.

LSP DB

After a PCC advertises LSP attributes of an MPLS network to all PCEs, each PCE stores these attributes in the label switched path (LSP) databases (DBs).

Stateful PCE

Stateful PCEs technique construct LSP DBs to monitor LSP information, including the assigned bandwidth and LSP establishment status, and use the LSP DB and TEDB information to calculate optimal paths for LSPs on an MPLS network.

Implementation

The PCE feature performs discovers PCEs. After members are discovered, PCCs and the PCE server establish PCEP sessions to exchange information. Before the ingress functioning as a PCC establishes an MPLS TE tunnel, the ingress sends a request to the selected PCE server to calculate a path and waits for the calculation result. Unlike IETF PCE, the NE20E allows you to verify and accept the calculated result or allows the PCE server to automatically confirm and accept the calculated path. After the calculated path is confirmed, the PCE server replies with this result to the PCC. Upon receipt the calculation result, the PCC establishes an LSP.

To improve network bandwidth usage efficiency and simplify network operation and maintenance, the NE20E implements Stateful PCEs and Uniform TE Network Information Configuration and Management.

PCE Discovery

An available PCC must be discovered before it sends a path calculation request to a PCE server. The PCE server, however, does not have to proactively discover a PCC. The NE20E only supports manually configured PCE member relationships. You need to specify the source IP address on a PCE server. The PCC then establishes a connection to the source IP address of the PCE server. You can specify multiple candidate PCE servers for the same PCC. The PCC selects a server based on the priority and source IP address. If candidate PCE servers have the same priority, the PCC selects a server with the smallest IP address. Other servers function as backup servers. If the server that is selected to calculate paths fails, the PCC automatically selects another server.

PCEP Sessions

The PCC establishes a PCEP session with a selected PCE within a specific domain, and the PCEs in different domains establish PCEP sessions with each other. The devices exchange information, including path calculation results, over the sessions.

Table 4-7 describes the implementation of a PCEP session.

Table 4-7 PCEP session implementation

Stage	Diagram	Description
PCEP session establishment	Figure 4-14 PCEP session establishment	The PCC initiates a TCP request to a PCE. After the PCC and PCE perform three handshakes, they establish a TCP connection. The PCC and PCE exchange Open messages to negotiate a session. After both devices accept one another's parameters, the two devices exchange Keepalive messages to confirm the negotiated parameters and use these parameters to establish a PCEP session.
PCEP session maintenance	Figure 4-15 PCEP session maintenance	A node on each end of the session periodically sends Keepalive messages to the other node to maintain the PCEP session. The transmissions on the two nodes are independent of each other. If one node fails to receive a Keepalive message after a specified interval time elapses, the node considers the session interrupted.
Disconnection	-	The node that fails to receive a Keepalive message sends a Close message to disconnect the PCEP session.

The process of establishing a PCEP session between two PCEs in different domains is similar to the preceding process of establishing a PCEP session between the PCC and PCE.

Intra-Domain Path Calculation

Intra-domain path calculation involves message exchanges only between the PCC and PCE. In the following example, the TEDB on each node contains the information about the entire network, and a session between the PCC and PCE has been established. Figure 4-16 illustrates intra-domain PCE path calculation. Table 4-8 describes intra-domain path calculation.

Figure 4-16 Intra-domain path calculation

Stateful PCEs

Stateful PCEs help establish optimal paths for both primary and backup TE LSPs. Although MPLS TE is used to properly assign network resources and improve network bandwidth usage, the TE LSP establishment mechanism insufficiently serves these purposes. In Figure 4-17, each link has 10 Gbit/s bandwidth. The LSP between nodes A and E needs 6 Gbit/s bandwidth, the LSP between nodes C and D needs 8 Gbit/s bandwidth, and the LSP between nodes C and G needs 4 Gbit/s bandwidth. The setup priority of the LSP between nodes A and E is the highest. The C-to-D link has less than 12 Gbit/s bandwidth and a priority lower than the A-to-E link. Without stateful PCEs, these three LSPs shown in Figure 4-17 (a) will be established. As a result, these established LSPs use all links on the network, which is an extremely inefficient use of network bandwidth.

Figure 4-17 LSP establishment with and without stateful PCEs

Alternatively, stateful PCEs can be used to improve network bandwidth usage. For example, in Figure 4-17 (b), stateful PCEs are used to establish the three LSPs over optimal paths. The bandwidth of the links between A and B, B and C, and D and E remain available.

Uniform TE Network Information Configuration and Management

The stateful PCE function enables a PCE server to uniformly configure and manage TE network topology and tunnel attributes, which streamlines network management and maintenance. The PCE server uses the configured TE topology information and tunnel attributes to calculate paths.