Path Computation Element (PCE) is a device that computes paths on behalf of the nodes in the network. It can be a router, a COTS server, part of the OSS, or a virtualized entity running in a cloud. When a network node needs a path for an LSP, it makes a request to the PCE using the PCE protocol (PCEP). The PCE has access to topology information for the entire network and uses this in path computations.
The PCE-based architecture and PCE protocol are defined by the IETF in RFCs 4655 and RFC 5440, respectively. PCEs provide an evolutionary path to Software Defined Networking (SDN).
Where does the PCE Server sit in the network?
A PCE Server provides a centralized path computation function. It sits between the OSS/NMS and NEs, as shown below.
The centralization of path computation enables operators to customize and control routing policies and algorithms from a single location. The successful and mature part of the architecture remains in the NEs. This architecture enables seamless inter-domain routing, flexible/customizable path computation, improved price/performance, and simplified operations.
PCE is the only technology allowing secure, dynamic, optimal, and private inter-area and inter domain TE path setup. Inter-domain routing is a winning application for PCEs and may alone be sufficient justification for deploying the PCE model in WANs/telco networks.
Customizable Path Computation
The PCE model enables network operators to customize the very path computation algorithms that are currently baked into router operating systems. The traditional path computation architecture gives operators little, or no, scope to radically change or to routinely increment, path computation. PCE server products provide open software APIs, to allow operators to customize or replace routing algorithms. These interfaces allow the OSS to influence network behavior via the PCE Server, in place of direct communications with every NE in the network. This is a much more reliable way to manage a network and cheaper for OSS vendors to develop and for OSS users to run. The increased flexibility and openness for customization enables operators to address the rapid change of pace set by today’s applications and traffic flows.
Improved Price / Performance Ratio
Deploying PCE requires the purchase of a small number of inexpensive PCE Servers and a one-time software upgrade of your existing equipment to support PCE (if it is not already supported). PCE Servers are not required to run on the data plane, therefore do not require expensive, proprietary hardware. Increasing path setup performance by adding and upgrading PCE Servers is much lower cost than adding or replacing data-plane NE.
Simplified Operations for Path Computation
Modifying or upgrading path computation policy no longer requires a change to all NEs in the network with the cost, time lag and instability that is involved in such a process. Instead a modification to a single PCE Server or small number of Servers is all that is required. For example, to upgrade the routing algorithm, the following steps could be taken (i) introduce a new Server with the new algorithm (ii) configure a few PCCs to talk to it and do some tests, (iii) once it is stable, cut over all the PCCs in the network to talk to the new Server. This very simple process replaces the upgrade of possibly thousands of NEs.
Hosting a PCE Server on a cluster of virtual servers in a cloud could provide even more flexibility, allowing for elastic use of resources (adding new virtual servers to the cloud at busy times) which reduces OPEX at less busy times.
This operational ease of upgrade supports the ability of the operator to customize the path computation function.