2011 was a great year for the networking industry with a major focus in the data center. This was a year of many announcements ranging from many topics/companies/technologies, including VXLAN, OTV,LISP, L2 DCI, and traffic trombones, Fabric Path, SPB/TRILL, 802.1BR, OpenFlow, SDN, Big Switch, Nicira, Embrane, scale out architectures, Hadoop and its impact on the network, virtual switching, virtual L4-L7 services, Fabric based architectures, 40GbE switching, tunnels, overlays, and the list goes on.
This was also the year I joined and started using twitter, just two months ago (and very glad I did), and found industry experts such as @ioshints, @bradhedlund, @etherealmind, @packetpushers, @ecbanks, @cloudtoad, @networkjanitor, and @scott_lowe just to name a few. Shortly after is when I started this blog.
As @networkjanitor said recently in a tweet to @bradhedlund, “As usual, another great post! You explained perfectly some of the random thoughts rattling around in my head,” and I could not agree more – that goes to the all of the blogs these guys write. Literally, in the past few weeks there have been many times I’m getting back from a customer or a normal day at work thinking about a technology/topic, and sure enough, one of these guys had just posted about it! Sometimes I think they are mind readers, but best of all, I appreciate their different perspectives on the various topics they post about.
Embrane officially launched this past Monday as you all are probably aware of already. By Monday evening, Brad
, and Greg
had already posted their take and impact of Embrane’s heleos platform. If you haven’t read them yet, I’d highly recommend doing so to get more acquainted with the solution. There are some great comments
on Brad’s post as well.
Prior to reading their write-ups, I did however, take a look at the whitepaper
, FAQ (focus on Licensing and Bill secton)
, and also watched the videos
by Dante Malagrinò on Embrane’s website.
What really caught my eye after going through the material was the Embrane Pricing Model. Based on my customer experiences (Enterprise to mid-market), heleos would do wonders for them, although they are NOT cloud service providers (CSPs), and that is who much of Embrane’s marketing material is targeted at…for now.
The Nexus 2000 as we all know gets its intelligence from its parent switch, i.e. Nexus 7000 or 5000. The Nexus 2000 for the most part is just a dumb device, otherwise known as a Fabric Extender (FEX). It may be a surprise for some, but even communication between two ports on the same fabric extender goes like this: FEX Source Port – Parent Switch – FEX Destination port. The same holds true for the Cisco UCS Fabric Interconnect (6100 and 6200 series) and the local FEX that resides in a UCS Blade Enclosure.
After having a recent conversation with Cisco and a financial customer around how L2 multicast works between a UCS 6120 and a FEX, it actually uncovered something interesting with regards to how this multicast flows.
Let me be fair by saying I work on designs on a regular basis that are 99% Cisco. Of course, there are integrations with other equipment for every technology, but from an R/S standpoint, it’s mainly Cisco. Occasionally, I’ll come across competitive (to CSCO) information, but that’s about it. For this post, I wanted to make it a point to see what was out there in the new 40 GbE LAN switching market. It’s a topic that is becoming more popular (for various trends in the Data Center) and I’m really quite surprised about it all, so I figured let’s dive in and see who’s got what.
And this was a shallow dive. The goal was not to spend countless hours on each solution; I simply wanted to get a high level overview. The focus was just to try and get the following questions answered.
Who has fixed configuration switches with 40GbE interfaces? Do they support “standard” L2/L3 protocols? Do they support some type of Layer 2 multipathing? Is there support for a type of MLAG? What is the port to port latency? What is the power consumption? Is there anything “special” about the switch, or is it unique in anyway?
After a few seconds of thinking about it, I decided to focus on Arista, Dell Force10, Extreme, IBM, Juniper, Brocade, HP, and of course, Cisco.