Comments inline...assume I am holding my breath, turning blue, and
kicking and screaming on the floor about certain terminology.
>From the Cisco website:
>
>http://www.cisco.com/warp/public/cc/so/cuso/epso/entdes/highd_wp.htm
>watch that wrap :->
>
>Layer 3 switching is hardware-based routing.
^^^^^^^
packet forwarding. packet forwarding. packet forwarding. packet forwarding.
NOT path determination or extensive processing
> In particular, the packet
>forwarding is handled by specialized hardware, usually ASICs. Depending on
>the protocols, interfaces, and features supported, Layer 3 switches can be
>used in place of routers in a campus design. Layer 3 switches that support
>standards-based packet header rewrite and time-to-live (TTL) decrement are
>called packet-by-packet Layer 3 switches.
>High-performance packet-by-packet Layer 3 switching is achieved in different
>ways. The Cisco 12000 Gigabit Switch Router (GSR) achieves wire-speed Layer
>3 switching with a crossbar switch matrix.
Wire speed, as with many vendors, is open to interpretation. Dig
through the release notes, and the Cisco-commissioned Tolly Group
reports, to see how much throughput drops unless one is EXTREMELY
careful with filtering and traffic shaping/policing.
>The Catalyst(r) family of
>multilayer switches performs Layer 3 switching with ASICs developed for the
>Supervisor Engine. Regardless of the underlying technology, Cisco's
>packet-by-packet Layer 3 switching implementations are standards-compliant
>and operate as a fast router to external devices
>
>End quote.
Without going into product futures, suffice it to say that I'm
actively involved in next-generation carrier router design. ASICs
have their role, but there are many specialized processing functions
that need different optimizations than a forwarding ASIC. DSP's.
Encryption. Compression.
And the path determination part of routing.
>
>But also look at this:
>
>[snip]
>Although higher-speed packet switching would seem to offer obvious benefits,
>a common but flawed assumption is that it will automatically improve the
>speed of applications. For example, if application demand on the network is
>10,000 pps, and current Layer 3 devices process packets at 200,000 pps,
>replacing them with devices that perform even at 10 million pps will not
>speed up the applications.
There is a very widespread misperception that routers and switches
impose significant latency. At one time, this was true. Be very
careful not to confuse port-to-port internal delays with queueing and
serialization delays. it takes, for example, 1200 microseconds for a
full-length Ethernet packet to get through an input interface at 10
Mbps.
___________________________________
UPDATED Posting Guidelines: http://www.groupstudy.com/list/guide.html
FAQ, list archives, and subscription info: http://www.groupstudy.com
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
