MPLS-based Request Routing

Current high-volume, high-availability data centers and content distribution networks use clusters of Web servers or caches to achieve scalability and reliability. In such environments a front-end dispatcher (or router) directs incoming client requests to one of the server machines. The request-routing decision can be based on a number of criteria, including server load, client request, or client identity. Dispatchers may be broadly categorized into two types: layer-4 dispatchers which base decisions on TCP and IP headers alone, and layer-7 dispatchers which use application layer information (e.g., HTTP headers) to direct clients. Content-based routing, for example, requires the dispatcher to terminate the incoming connection, examine the higher-layer headers, and create a new connection with the appropriate server (or transfer the incoming connection). As Web application requirements evolve, there will be a need for more sophisticated request routing, which implies that examining only transport-layer headers provides insufficient functionality. But layer-7 dispatchers, while more sophisticated, suffer from limited scalability and performance since they must perform connection termination and management for a large number of clients. Ideally, a dispatcher should provide the flexibility of layer-7 forwarding, while maintaining performance levels comparable with layer-4 hardware switches. In this synopsis we describe an approach that, when combined with an intelligent client-side proxy, can implement a dispatcher using commercial, high-performance, off-theshelf switching hardware, while also providing the flexibility of a content-based router.