Use seperate CSU/DSU or WIC T1 CSU/DSU for WAN link?

Unanswered Question
May 24th, 2008
User Badges:

Forgive me but I just had a basic question about hardware.

Our network was designed and implemented (before I got here) by a CCIE.

All branches have MPLS connectivity and 2800 series routers.

At each site, he used a seperate CSU/DSU.

I don't know why he did that when you can have a WIC T1 CSU/DSU for a cleaner connection and one less thing to supply power to.

Is there any technical reason to do that that I am unaware of that makes a better or safer or more efficient connection?

  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 5 (2 ratings)
paolo bevilacqua Sat, 05/24/2008 - 08:26
User Badges:
  • Super Gold, 25000 points or more
  • Hall of Fame,

    Founding Member

There is really no technical reason and is better to use integrated CSU/DSU for the very reasons you mentioned.

Hope this helps, please rate post if it does!

Richard Burts Sat, 05/24/2008 - 11:40
User Badges:
  • Super Silver, 17500 points or more
  • Hall of Fame,

    Founding Member

  • Cisco Designated VIP,

    2017 LAN, WAN


I agree with Paolo that there is not much technical reason to prefer separate CSU/DSU. But is it possible that there was some financial or political reason to choose this (layers 8 and 9 of the multi layered model)? Is it possible that the circuits with separate CSU/DSU existed with older routers and the routers were upgraded (in which case it probably was less expensive to keep the existing CSU/DSU and use a less expensive serial card)? Is it possible that in negotiating for provisioning of the circuits the provider gave an especially attractive price for circuits with CSU/DSU included? Is it possible that in setting up Service Level Agreements where responsibility for the performance of the serial links goes to the provider that they wanted/needed control of the CSU/DSU?




This Discussion