Cisco Hierarchy Archittecture : Dual Core Design

[vtandon]

Hello friends,

Can somebody explain with the design aspects considered when design the LAN with dual and collpased core topology.

basically i need to know why we are adding core layer in the topology which can add the delay/latency in the network.

regds

VT

[ankurbhasin]

Hi VT,

See it all depends on how big your network is.

Having a small network you have your routing and switching done on same layer and can say it as a collapse layer by whcih I mean connecting you user switch to the distribution switch and doing the layer 3 (intervlan routing and other routing) and switching on same switch it will definetely increase latency if you have avery big network and lot of server farms and internet connections.

Having a very large network full of server farms , mainframes , and many internet connections it is better to have dual layer where you can connect all your servers and other stuff to core so that switching to it can be very fast (hardware switching)and need not waste time looking into routing tables and other stuff. With that kind of network the latency which you will see will be quite less as compared to which it will have on a collapse layer for big netwirk.

Truely to say its all conceptual and depends on your network and network requirement and yes your network size also decides this.

HTH

Ankur

[vtandon]

Hi ankur,

Thanks for reply .

But as per the Cisco. Dual core design is used to connect two or more switch block ( i.e multipal access switch & Two dist. switch/ L3 switch ) . In this senario how it is useful.

Regards,

Vijay Tandon.

[vtandon]

Hi,

Pls go thtouch below mentioned URL .i want to know what are the advatage of dual core L2 swithc ..

http://www.cisco.com/en/US/netsol/ns340/ns394/ns147/ns17/networking_solutions_design_guidance09186a00800a3869.html

Regards,

Vijay Tandon

[Josef Oduwo]

According to the link you have posted, the main point of having a dual L2 switched backbone is "cost-effective performance and high availability".

According to Cisco recommendations (DESGN course content - http://www.cisco.com/pcgi-bin/front.x/wwtraining/CELC/index.cgi?action=CourseDesc&COURSE_ID=4418), the Core's functions are "fast transport, high reliability, Redundancy, Fault tolerance, low latency, manageability et al. These two ideas are effectively the same and pretty self explanatory. Please note that Core functions are predominantly L2 whilst Distribution layer functions are both L2 and L3 (ACLs, QoS, Security etc).

There are two separate and somewhat conflicting ideas here: one is if you have a server farm which needs an extra layer/level of redundancy and fault tolerance and secondly, a case where all is required is fast switching. The former is the Enterprise Composite Network Model (a mess of "layers") applies and the latter is the Hierarchical Network Design model (with three layers) situ.

The Hierarchical model can be broken up even further into the Switched or Routed "sub-models". Which is where your question comes in: in the Switched Hierarchical Design model, the dual core mainly provides fast transport. Routing functionality is in the Distribution layer. Otherwise have a browse at CCDA/DESGN above and design books for more info.

Josef.

PS - please rate if it 'elps...