Routing in Campus Area Network - Three layer Architecture

[Dipesh Patel]

Dear All Experts,

We are planning to upgrade out network infrastructure.

Current Architecture :

• Three Layer Architecture --- CORE to Distribution to Access : All connected via switched links.
  
• Core switch - Only one
  
• Distribution Switches - 18 no's
  
• Access Switches - 500 no's
  
• Static routing between Distribution to Core and iBGP between CORE to MPLS CE router.
  
• Serverfarm Vlan is on CORE switches only
  

Proposed Architecture :

• Three Layer Architecture --- CORE to Distribution : All connected via dual uplinks from two nearest CORE switches with Routed links. CORE to CORE also routed links
  
• Core switch - 4 no's
  
• Distribution Switches - 18 no's
  
• Access Switches - 500 no's
  
• OSPF routing protocol between CORE to Distrbution and iBGP between CORE to CORE and CORE to MPLS CE router.
  
• Serverfarm Vlans will be on core switches only and connected to CORE -1 and CORE - 2 using HSRP
  

Can you please help to design OSPF network?

Proposed Architecture is inserted below :

1. How Area should be defined or should I go for same Area 0 for all distribution?

2. How OSPF will chose the best path to CORE?

3. As per OSPF SPF algorithm if it will chose CORE - 2 as the shortest path and using iBGP i have chosen CORE - 1 is the primary than how it will work?

Regards

[Joseph W. Doherty]

Disclaimer

The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.

Liability Disclaimer

In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.

Posting

1. How Area should be defined or should I go for same Area 0 for all distribution?
2. How OSPF will chose the best path to CORE?
3. As per OSPF SPF algorithm if it will chose CORE - 2 as the shortest path and using iBGP i have chosen CORE - 1 is the primary than how it will work?

#1 So about 25 OSPF devices?  One area would likely suffice.  It could be area 0 or a numbered area.

#2 Lowest OSPF past cost.

#3 Default ADs will use OSFP in preference to iBGP.

[Dipesh Patel]

Dear Joseph,

Can you reply how complete topology should be configured?

[Joseph W. Doherty]

Disclaimer

The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.

Liability Disclaimer

In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.

Posting

What you might consider is running OSPF on all your internal L3 devices (this in addition running BGP now).

If your servers are directly connected to cores, avoid spanning server VLANs across core devices.  (To avoid unicast flooding.)

[Dipesh Patel]

Dear JosephDoherty,

The complete idea is ,

At present iBGP is there between MPLS CE routers and  one CORE switch.

Now we are planning to install 3 more core switches.

Also we are planning to configure OSPF between CORE to Distribution and iBGP between CORE to CORE.

1. Can you sugguest how can I achieve?

Regards

[Joseph W. Doherty]

Disclaimer

The  Author of this posting offers the information contained within this  posting without consideration and with the reader's understanding that  there's no implied or expressed suitability or fitness for any purpose.  Information provided is for informational purposes only and should not  be construed as rendering professional advice of any kind. Usage of this  posting's information is solely at reader's own risk.

Liability Disclaimer

In  no event shall Author be liable for any damages whatsoever (including,  without limitation, damages for loss of use, data or profit) arising out  of the use or inability to use the posting's information even if Author  has been advised of the possibility of such damage.

Posting

1. Can you sugguest how can I achieve?

What I suggested was extending OSPF to include the links between the core devices too.

You might also include the core to CE links in OSPF.

Including these additional links and routers (CEs) in OSPF insures the whole site topology is contained within your IGP (OSPF).  BGP is used just to get to/from your WAN clouds.  Every core knows, via BGP, best CE for off-site destination traffic.  CEs (if they too participate in OSPF) know best internal path for all received external traffic.

If you extend OSPF to CEs, you also have the option to limit BGP just to them (i.e. remove from cores).  However, this will require some kind of redistribution of external routes, perhaps aggregated, into your OSPF topology.

Also depending on your CE platforms, OER/PfR might be run on them to provide optimal egress path selection.