Arista launches the first hardware VXLAN termination device

Arista is launching a new product line today shrouded in mists of SDN and cloud buzzwords: the 7150 series top-of-rack switches. As expected, the switches offer up to 64 10GE ports with wire speed L2 and L3 forwarding and 400 nanosecond(!) latency.

Also expected from Arista: unexpected creativity. Instead of providing a 40GE port on the switch that can be split into four 10GE ports with a breakout cable (like everyone else is doing), these switches group four physical 10GE SFP+ ports into a native 40GE (not 4x10GE LAG) interface.

But wait, there’s more...

The 7150 switches are the first devices that offer VXLAN termination in hardware. Broadcom’s upcoming Trident-2 chipset supports VXLAN and NVGRE, so when Arista demonstrated VXLAN termination at the recent VMworld 2012, everyone expected the product to be available next spring ... but according to Arista it’s orderable now and shipping in Q4. Turns out Arista decided to use Intel’s chipset this time, proving yet again that they can be remarkably agile with regard to the merchant silicon.

Another goodie: you can run IEEE 1588 (Precision Time Protocol) on these devices to establish an extremely precise time base in your network, drifting only a few nanoseconds per day (precision clock module seems to be optional). Such a precision might not make sense at the first glance (unless you’re working in high-frequency trading), until you discover you can timestamp mirrored (Arista’s name for SPAN) or sFlow packets. Imagine being able to collect packets across the whole network and having a (almost) totally reliable timestamp attached to all of them.

Finally (and my friend Tom Hollingsworth will love this part), 7150 switches can do NAT in hardware. Yeah, you got that right – they do NAT in silicon (don’t even try to ask me whether it’s NAT44, NAT64, or NAT66 ;) with less than one microsecond latency.

Update 2012-09-20: It seems 7150 will only be able to do NAT44 - that's all that Intel's FM6000 chipset supports.

Does all of this sound too good to be true? I learned the hard way never to trust product launches and datasheets; let’s wait for the new EOS documentation to become available – after that, we’ll see how well the actual features match the promises.

6 comments:

  1. http://www.lightreading.com/document.asp?doc_id=225068

    Cisco's Nexus 3548, announced Wednesday morning, claims latency of 190 nanoseconds through the switch, compared with Arista's 350 nanoseconds.
    The 3548 is the switch Cisco was supposedly marketing as an "Arista killer." And it got announced on the same day that Arista unveiled its 350-nanosecond latency switches, the new 7150 series.
    Replies
    1. Yeah, Arista and Cisco launching competing products on the same day must be a pure coincidence ;) More details on Nexus 3548 here: http://blog.ioshints.info/2012/09/cisco-nexus-3548-victory-for-custom.html
  2. According to intel one of the usage for the NAT will be SLB on the switch

    http://communities.intel.com/community/wired/blog/2012/09/04/supporting-data-center-load-balancing-at-the-switch

    you can read more about the intel FM6000 here
    http://www.intel.com/content/dam/www/public/us/en/documents/white-papers/ethernet-switch-fm6000-sdn-paper.pdf

    Nitzan



  3. Hi Ivan,

    With 12.5 Brocade ADX L4-L7 switches can act as VTEPs. It has been announced on Aug. It is supported on existing product line of ADX series.

    http://newsroom.brocade.com/press-releases/brocade-adx-series-to-unveil-vxlan-gateway-and-app-nasdaq-brcd-0923542
    Replies
    1. Thanks for the comment. Checked Brocade web site, 12.5 documentation is not yet out there. Once it ships, please send me an email.
  4. WHat is that Intel chipset called ?
Add comment
Sidebar