Delving in Cisco Data Center Fabric, Part 1

March 25, 2011 - 1 Comment

Following my last blog post , I have gotten a number of questions on how we specifically define “fabric” so I thought I’d dig into that a bit more with this post.  So, the primarily point is that our definition of fabric it built around a specific set of features and capabilities. It is not tied to specific products or topology. Again, we think it’s important that our customers have choice and not have an arbitrary architecture foisted upon them.

At its most basic level, a fabric is a highly available, high performance shared infrastructure built with integrated, intelligent compute, storage and network nodes that can be rapidly and simply organized around the requirements of a given workload.

We see this fabric as having six specific characteristics:

    • Open – based on open standards
    • Integrated – breaks down traditional silos with a more holistic approach
    • Flexible – allows customers architectural flexibility and choice
    • Scalability – easily grows and adapts as the data center evolves
    • Resilient – delivers bullet-proof uptime
    • Secure – provides security and policy compliance

Today, let’s talk about “open” – I’ll address the other characteristics in subsequent posts…

For a fabric to be effective it must be pervasive and for it to be pervasive, it must be based on open standards.  Data centers are inherently heterogonous and the only way to build an inclusive, ubiquitous environment is to base it on open standards.  In our case, our data center fabric is either build with established standards (10GbE, Fibre Channel, FCoE), pre-standards with a commitment to support ratified standards (IEEE DCB, FabricPath/IEEE TRILL) or standards-based alternatives to Cisco innovation (VPLS for data center interconnect). This approach gives you the choice and the control of the vendors you work with. For example, you can incorporate non-UCS servers into our fabric solution and not lose functionality such as LAN/SAN convergence or OTV-based data center interconnect.

So, this is our take on an open fabric and why its important.  What do you think?

BTW please join me on Wednesday March 30th at 9AM PT for our event on Data Center Fabric.

In an effort to keep conversations fresh, Cisco Blogs closes comments after 60 days. Please visit the Cisco Blogs hub page for the latest content.


  1. Thanks for the concise explanation of ‘fabric’ — it will be easier to explain this to customers with this language. One thing, though: I think you mean “heterogeneous” and not “heterogonous”, although likening a data center to a garden is an interesting analogy 😉