One of the fun things for me about helping with the new CCDP Bootcamp is learning about technologies you do not get exposure to in the “traditional” Routing and Switching curriculum that are based on mid-level and low-level Cisco hardware. One of those technologies that our new course will cover is Bidirectional Forwarding Detection (BFD). This technology is found on Cisco 6500/7600 series routers, as well as 12000 series and Carrier Routing System (CRS-1 Routers).

Like many of the features covered in the Routing Protocol Design section of the course, Bidirectional Forwarding Detection seeks to speed up routing protocol convergence. The key factor in this goal addressed by BFD is the fast detection of node or link failure in the routing design.

Many approaches to the fast detection of the failure of a link or neighbor deal with some type of accelerated hello packet (fast hellos at Layer 3). BFD operates along a similar line,except the fast link hellos are accomplished at Layer 2. The CPU impact of the BFD process ends up being much less compared to other fast hello approaches. Cisco has tested systems running 100 concurrent BFD sessions between routers and noticed an increase in CPU utilization of a mere 2%. And remember, that was with 100 concurrent sessions running.

In order to take advantage of Cisco’s BFD, we not only require the correct software and hardware version, but we need the cooperation of the routing protocol. Fortunately, BFD functionality has been engineered into OSPF, EIGRP, IS-IS, and BGP, as well as such important “support” protocols like HSRP.

If you would like to learn more about BFD right now, visit the Cisco DOC-CD.

You can leave a response, or trackback from your own site.

6 Responses to “Bidirectional Forwarding Detection (BFD)”

  1. We used BFD in a large(-ish) multi-domain MPLS network (which used Cat6500s in the core and aggregation layers) in the UK some 4-5 years ago, and it was simple to understand and configure: it just worked.

    These days, of course, BFD is widely supported with other vendors as well (e.g. Juniper and Alcatel-Lucent).

  2. zeus says:


    r u teaching this class?

  3. Nikita says:

    “BFD operates along a similar line,except the fast link hellos are accomplished at Layer 2.”
    As far as I remember BFD uses UDP as underlying protocol. The main advantage of the BFD is that this hello packets proccessed on a line card’s hardware and doesnt go all the way to the Routing Engine(Processor)

  4. Pete says:

    One of the biggest improvements BFD brings to the table, besides being incredibly lightweight, is that Cisco devices handle it in the interrupt context, unlike a fast hello that will done as part of the scheduling for that specific process.

    Although only on the 12k today, BFD will most likely find itself being pushed down to the hardware on the 6k/7600 as well and operating in a distributed manner and never actually hitting the CPU.

  5. waseem says:

    plz tell me that bfd works at which layer???


Leave a Reply


CCIE Bloggers