Overview of the Cisco 12008
Cisco 12008 Gigabit Switch Router Installation and Configuration Guide
1-68
Once the forwarding decision has been made, the silicon queuing engine is notified by
the forwarding processor, and the silicon queuing engine places the packet in the proper
queue.
This partitioning between the Layer 2 switching accelerator and the forwarding
processor blends the high throughput of hardware-accelerated forwarding with the
flexibility of software-based routing.
•Silicon queuing engine—Each line card has two silicon queuing engines: receive and
transmit. The receive engine moves packets from the burst buffer to the switch fabric,
and the transmit engine moves packets from the switch fabric to the transmit interface.
When an incoming IP packet is clocked into the silicon queuing engine, the packet’s
integrity is verified by a CRC check. Next, the silicon queuing engine transfers the IP
packet to buffer memory and tells the Layer3 switching accelerator the location of the
IP packet. Simultaneously, the silicon queuing engine is receiving forwarding
information from the forwarding processor, while the forwarding processor is telling the
silicon queuing engine where the IP packet is to be placed in the virtual output queue.
Each virtual output queue represents an output destination (destination line card).
Placement of the IP packets in a virtual output queue is based on the decision made by
the forwarding processor. There is one virtual output queue for each line card, plus a
dedicated virtual output queue for multicast service.
The transmit silicon queuing engine moves the packet from the switch fabric to the
transmit buffer, and then to the transmit interface.
•Switch fabric interface—The switch fabric interface is the same 1.25-Gbps, full-duplex
data path to the switching fabric that is used by the GRP. Once a packet is in the proper
queue, the switch fabric interface issues a request to the master clock scheduler on the
CSC. The scheduler issues a grant and transfers the packet across the switching fabric.
•Maintenance bus (MBus) module—An MBus module on the line card responds to
requests from the master MBus module on the GRP. The line card MBus modu le reports
temperature and voltage information to the master MBus module.
In addition, the MBus module on the line card contains the ID-EEPROM, which stores
the serial number, hardware revision level, and other information about the card.
•Cisco Express Forwarding (CEF) memory table—Each line card maintains CEF tables.
These tables, derived from routing tables maintained by the GRP, are used by the line
card processor to make forwarding decisions.