CORE 6509 design question

Unanswered Question
Jun 14th, 2007
User Badges:

In the Cisco Data Center Infrastructure 2.1 SRND (link below), this has been stated on page 2-3:

The data center core is interconnected with both the campus core and aggregation layer in a redundant

fashion with Layer 3 10 GigE links. This provides for a fully redundant architecture and eliminates a

single core node from being a single point of failure. This also permits the core nodes to be deployed

with only a single supervisor module.

For a client redesign, I originally had planned out a dual-6509 dual-supervisor model at the condensed distribution/core layer and 3750's at the access layer. But I am confused - is that actually needed? Is the redundancy and resiliency between a dual 6509 dual-sup vs. dual 6509 only very marginal? I prefer technical explanations to just telling the customer that he would really me more redundant with buying more product.

Also, the SRND does seem to suggest that 'it's okay' going with two 6509's with single sup's in each. Unless I am reading their language wrong.

As for the client's requirement, they do want redundancy in case a supervisor or chassis fails but what's the likelihood of both happening at the same time?

Thank you

  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 0 (0 ratings)
Jon Marshall Thu, 06/14/2007 - 22:34
User Badges:
  • Super Blue, 32500 points or more
  • Hall of Fame,

    Founding Member

  • Cisco Designated VIP,

    2017 LAN, WAN


We use single supervisors in our core/distribtion switches in our data centre. We can do this because we insist that anything that connects into these switches must be dual-honed to both switches. There are other parts of our network where we have deployed dual-sup confgurations because we cannot ensure that all connections into the switches are dual-honed. That is our main criteria.

Sup's can and do fail but as long as you have an alternate path you should be okay. For maximum redundancy go with dual sups but as with all design it often comes down to cost.




This Discussion