Dell MXL/IOA with FC Flex IOM: Network and Infrastructure Convergence with NPIV Proxy Gateway

Dell FC Flex IO Module for Dell MXL/IOA Blade Switch

Dell FC Flex IO Module for Dell MXL/IOA Blade Switch


Very exciting news from Dell today as they announced the availability of the Dell FC Flex IOM module which transforms the Dell M1000e MXL/IOA blade switch into a NPIV Proxy Gateway (NPG). This allows the MXL/IOA to bridge between Ethernet and Fibre Channel allowing for network convergence with FCoE.


The Dell FC Flex IOM is a 4-port FC (2/4/8G) module that can be inserted into one or both of the Flex I/O slots on the Dell MXL/IOA blade switch for the Dell M1000e chassis. The FC Flex IOM allows for the MXL/IOA to bridge between Ethernet and Fibre Channel and acts as a NPIV Proxy Gateway.


The Dell S5000, Cisco Nexus 5548UP, or Brocade VDX line are all converged/FCoE switches that provide this Ethernet-FibreChannel bridging capability at the ToR layer – the Dell MXL/IOA w/ FC Flex IOM brings the convergence layer down to the blade level. It provides not only network convergence but also infrastructure convergence and consolidation. The below network diagram displays such a topology.

Converged Network with Dell MXL and FC Flex IO Module

Converged Network with Dell MXL and FC Flex IO Module


The great thing about the Dell MXL/IOA blade switch w/ FC Flex IOM is that since it acts as a NPIV Proxy Gateway, it does not consume a Fibre Channel Domain ID and any back-end existing FC SAN infrastructure remains in place allowing for an easy transition to network convergence with FCoE.


Additionally, since the Dell MXL/IOA blade switches have 2 x Flex I/O slots, it is possible to utilize 2 x FC Flex IOMs for a total of 8 x FC ports per blade switch. It is also possible to use the FC Flex IOM with other FC Flex I/O moudles such as the 4-port 10 GbE SFP+ module, the 4-port 10GBaseT module, or the 2-port 40 GbE module.


Check-out my latest Dell Networking FC Flex IOM whitepapers on dell.com and converge your network today! :-)


Dell FC Flex IOM Whitepapers:


Dell Networking FC Flex IOM: Infrastructure & Network Convergence w/ FCoE


FC Flex IOM Deployment of FCoE with Dell FC Flex IOM Brocade FC Switches and Dell Compellent Storage Array


Deploying Dell Networking MXL and PowerEdge M IO Aggregator with the FC FlexIO Module in a Cisco MDS Environment


Twitt

Tags: , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,

5 Responses to “Dell MXL/IOA with FC Flex IOM: Network and Infrastructure Convergence with NPIV Proxy Gateway”

  1. Bobby says:

    Your white paper “Dell Networking FC Flex IOM: Deployment of FCoE with Dell FC Flex IOM, Brocade FC switches, and Dell Compellent Storage Array” Figure 5 shows multiple fibre channel links (fi 0/41-44) between the MXL FC Flex IOM ports and the SAN switch ports. How do you configure the MXL FC Flex IOM so that fibre channel traffic will run on more than one of these links at once? It seems like I/O will only run on one at a time unless you somehow configure trunking.

  2. Humair says:

    Hi Bobby,

    FC switch vendors do not support connecting FC ports (E-ports) to other vendor FC switches for the creation of trunk ports. For example, in Full Fabric Mode or FCF mode, you cannot connect FC ports from Cisco and Brocade together as Cisco and Brocade FC switches will not interoperate in this mode.

    The Dell MXL/IOA blade switch w/ FC Flex IOM runs in NPIV Proxy Gateway mode. This ‘gateway’ mode supported by Dell and other vendors (Cisco calls it ‘NPV’ and Brocade calls it ‘Access Gateway’) allows for interoperability between different vendor switches as the converged switch simply bridges between Ethernet and FC but never consumes a Domain ID or becomes part of the actual FC fabric.

    As such, for interoperability, this ‘gateway’ mode can be used, but there is no support to create what is known as F-port Trunking between different vendor switches whether Dell, Cisco, or Brocade.

    When an ENode logs into the fabric, the Dell MXL/IOA blade switch w/ FC Flex IOM will use round robin methodology for the logins. So, I can still connect all four FC ports on the FC Flex IOM (or 8 FC ports with 2 x FC Flex IOMs) to a backend FC switch. As logins occur from ENodes, the logins will round robin over the FC ports.

  3. Bobby says:

    Thank you for the explanation!

  4. Yani says:

    I get that trunking is not going to work but If I have more than one Link from the FC module on the MXL to say a Brocade 300, will it behave as a redundant link. (i.e. if I pull out on of those cables (doesn’t matter which one) should the other one continue to operate?

  5. Humair says:

    Typically, you would have two isolated fabrics – Fabric A for CNA port 1 and Fabric B for CNA port 2. A failure on one Fabric would allow the server to keep accessing storage via the other Fabric.

    Now in the case you mention, if an uplink FC port fails on Fabric A and an ENode has already logged in over that respective FC port on the FC Flex IOM, the ENode should login over any other available port on the same FC Flex IOM connecting to the Brocade 300. Of course the same Round Robin load balancing methodology will be utilized.

Leave a Reply

*


9 − = seven