Layer 3 switching is routing to all intents and purposes, it is a L3 decision made on where to send a packet.
Layer 3 switching was used largely as a marketing term at one time because everyone knew switching was fast and routing slow because switching was done with dedicated hardware and routing was done on a router in software.
Along came the L3 switch and suddenly routing could be done in hardware as well.
Thanks for your reply Jon,
My focal point is , if I've for example 3750 + EMI with 20 hosts in my LAN (forget about security) is that good enough to communicate with External world? If so, how the link termination from ISP.
could anyone dig it more.
Yes it is enough providing that the presentation from the ISP connection is fibre or RJ45.
If it is then you could configure the connection between your 3750 and the ISP router as a P2P /30 routed link eg.
ip address 192.168.5.1 255.255.255.252
ISP end would be 192.168.5.2/30.
Then add a defult route on your 3750
ip route 0.0.0.0 0.0.0.0 192.168.5.2
Only other thing is NAT. You meay need to NAT your internal client addresses - depends on whether your or the ISP is natting.
Thanks for u r explanation Jon,
Then why so many companys introducing the router between LAN and ISP. wht is the adv and disadv ?
Correct me if I'm wrong
Adv : Cost Saving ( Without Router)
Dis Adv : NAT and VPN
Thanks in Advance.
There may be a number of reasons
1) The termination of the ISP is not presented as ethernet and routers generally support a lot more types of termination.
2) The concept of layered security is important. You would not really want to expose your internal switch to the outside world. Note that with the 6500 switch and a Firewall Services Module (FWSM) you could do this.
3) NAT as you say. Only the 6500 switch supports NAT but pretty much all routers do.
4) Again as you VPN's.
L3 switches, 6500 aside which are an entirely different subject, are good for internal LAN connectivity and inter-vlan routing. That is where they are most likely to be deloyed.
Hope this makes sense
As Jon has already noted, L3 switching usually implies hardware supported routing. The hardware often delivering L2 switching performance.
For instance, a 7200 (very high end pure) router with G2 processor has a 2 Mpps rating, yet the 3560G/3750G (low end) L3 switches support up to 38.7 Mpps.
What routers lack in performance, though, they usually make up for in many other features and additional interface types support (also touched upon by Jon).
Hi Ravi A,
As per Cisco:
It is important to understand the difference between Layer 3 routing and Layer 3 switching. Both terms are open to some interpretation; however, the distinction between both can perhaps be best explained by examining how an IP packet is routed. The process of routing an IP packet can be divided into two distinct processes:
Â·Control plane-The control plane process is responsible for building and maintaining the IP routing table, which defines where an IP packet should be routed to based upon the destination address of the packet, which is defined in terms of a next hop IP address and the egress interface that the next hop is reachable from. Layer 3 routing generally refers to control plane operations.
Â·Data plane-The data plane process is responsible for actually routing an IP packet, based upon information learned by the control plane. Whereas the control plane defines where an IP packet should be routed to, the data plane defines exactly how an IP packet should be routed. This information includes the underlying Layer 2 addressing required for the IP packet so that it reaches the next hop destination, as well as other operations required on for IP routing, such as decrementing the time-to-live (TTL) field and recomputing the IP header checksum. Layer 3 switching generally refers to data plane operations.
Thanks Goutam (Please rate if it helps you)