Greg Minshall

IP switching and gigabit routers

To cope with the growth in the Internet and corporate IP networks, we require IP routers capable of much higher performance than is possible with existing architectures. This article examines two approaches to the design of a high-performance router, the gigabit router and the IP switch, and then provides some detail on the implementation of an IP switch and the protocols associated with IP switching.

IP switching—ATM under IP

Internet protocol (IP) traffic on the Internet and private enterprise networks has been growing exponentially for some time. This growth is beginning to stress the traditional processor-based design of current-day routers. Switching technology offers much higher aggregate bandwidth, but presently only offers a layer-2 bridging solution. Various proposals are under way to support IP routing over an asynchronous transfer mode (ATM) network. However, these proposals hide the real network topology from the IP layer by treating the data-link layer as a large opaque network cloud. We argue that this leads to complexity, inefficiency, and duplication of functionality in the resulting network. We propose an alternative in which we discard the end-to-end ATM connection and integrate fast ATM hardware directly with IP, preserving the connectionless nature of IP. We use the soft-state in the ATM hardware to cache the IP forwarding decision. This enables further traffic on the same IP flow to be switched by the ATM hardware rather than forwarded by IP software. We claim that this approach combines the simplicity, scalability, and robustness of IP, with the speed, capacity, and multiservice traffic capabilities of ATM.

Flow labelled IP over ATM: design and rationale

We describe a system in which layer 2 switching is placed directly under the control of layer 3 routing protocols on a hop-by-hop basis. Specifically, ATM switching is controlled by IP. We couple each ATM switch with a general purpose computer running IP routing and management protocols. We define a default ATM virtual channel identifier (VCI) to be used for transmitting IP packets over ATM links. We then define mechanisms which allow specific flows to be transmitted on specific ATM VCIs. The resulting system obeys IPs semantics for routing and forwarding, and takes advantage of ATMs switching hardware to accelerate the forwarding of packets. While this system takes advantage of ATM hardware, the ATM signalling, routing, and management architecture (as specified by the ATM Forum) is replaced by the protocols and practices currently in use for IP routing and management.