switching methods

carl_townshend · ‎06-09-2009

Hi all

I am a little confused with the difference in the switching methods. can anyone explain very simply the below, I am confused mainly with the fast switching vs cef, but please explain all.

1.Normal switching method

2.fast switching

3.cef

is there much difference in all on these?

francisco_1 · ‎06-09-2009

take a look http://books.google.co.uk/books?id=dKzpj4r7KCwC&pg=PA178&lpg=PA178&dq=cisco+fast+switching,+cef&source=bl&ots=qMsRW4VQ3M&sig=SHeNnvA60cKG9xkwJsVKZIBNTNI&hl=en&ei=biUuSu7NDI3MjAehu6iYCw&sa=X&oi=book_result&ct=result&resnum=4#PPA178,M1

sorry about the long URL but book is worth reading...

networker.cisco · ‎06-09-2009

hi,

http://www.cisco.com/en/US/tech/tk827/tk831/technologies_white_paper09186a00800a62d9.shtml

above link also help for your understanding

rgds/shiva

Manish Prasad · ‎06-09-2009

Hello CARL,

This might helpâ€¦â€¦.

1) Normal Switching or Process sw:-It will look up in the routing table for each packet destination ip and then search for the next hop ip address.

2) Fast sw:-In fast switching the very first packet will be processed with the Normal switching technology i.e. process switching and then the route is cached, keep on looking the sequence of the packets, if it is next in sequence then it processed through cache. It doesn't have to look up in routing table for each packet except the first packet.

3) CEF:-In CEF the first packet problem is solved. It creates a separate table from routing table called FIB (Forward Information Base). It will no longer look up the Routing table and get the packet processed by FIB. With FIB one more table generated i.e. ADJ table (adjacency table) and it contains next hop ip address .Through CEF we can do per packet load sharing. Using CEF might eat up some extra memory to handle the FIB and ADJ tables.

Thanks

Prasad