5.3 Tag Encapsulation on Non-ATM Links

5.3 Tag Encapsulation on Non-ATM Links

With some link layer technologies (e.g., ATM), the link layer header has adequate semantics to carry tag information. For such technologies, carrying tag information in the link layer header has an advantage of allowing the re-use of existing forwarding functionality (e.g., the ability to use ATM switch hardware). However, there are other link layer technologies (e.g., point-to-point links, Ethernet, FDDI, Token Ring, etc.), whose link layer headers don?t have semantics adequate to carry tag information. When Tag Switching is used over subnetworks built out of such technologies, tag information is carried in a small ?shim? inserted between the link layer and the network layer headers.

On point-to-point subnetworks, packets that carry tags are identified by the PPP protocol field, with one value used to identify unicast and another to identify multicast. On multiaccess subnetworks such packets are identified by the ethertype. Like point-to-point subnetworks, multiaccess subnetworks use one ethertype to identify unicast and another to identify multicast.

To support the notion of a tag stack as discussed in Section 5.1.2 , the shim layer used by Tag Switching consists of a sequence of tag stack entries, where the top of the tag stack appears in a packet right after the link layer header, and the bottom of the tag stack appears in the packet right before the network layer header. Packet forwarding is determined by the entry at the top of the stack. Each tag stack entry is encoded as shown in Figure...

< Previous Excerpt Next Excerpt >

Purchase This Book

Switching in IP Networks: IP Switching, Tag Switching, & Related Technologies

TABLE OF CONTENTS

5.3 Tag Encapsulation on Non-ATM Links

Contact Preferences

This is embarrasing...

Customize Your GlobalSpec Experience

Select Your Free Newsletters

Industry Newsletters

Select Your Free Product Alerts

This is embarrasing...