Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!


Forgot your password?

Inventor of OpenFlow SDN Admits Most SDN Today Is Hype 62

darthcamaro writes "Every networking vendor today is talking about Software Defined Networking (SDN). The basic idea is that the control of the underlying networking hardware is abstracted by software. Martin Casado helped to come up with the whole topic with his 2005 Stanford thesis (PDF). Eight years later after selling his startup Nicira to VMware for $1.2 Billion, Casado sees the term SDN meaning everything and nothing to all people. From the article: '"I actually don't know what SDN means anymore, to be honest," Casado said. Casado noted that the term SDN was coined in 2009 and at the time it did mean something fairly specific. "Now it is just being used as a general term for networking, like all networking is SDN," Casado said. "SDN is now just an umbrella term for, cool stuff in networking."'"
This discussion has been archived. No new comments can be posted.

Inventor of OpenFlow SDN Admits Most SDN Today Is Hype

Comments Filter:
  • by GeneralTurgidson ( 2464452 ) on Monday April 29, 2013 @08:29PM (#43586553)
    I need to build a business around some new buzzword and sell it to VMware. Cloud and everything related to it has really stagnated development of other areas of IT in my opinion. Companies try and figure out WTF SDN is or how to integrate their networking stack with AWS instead of focusing on what's really happening in the IT world.
  • by Anonymous Coward on Monday April 29, 2013 @08:29PM (#43586555)

    It's the way of the future.

  • by John Hasler ( 414242 ) on Monday April 29, 2013 @08:30PM (#43586559) Homepage

    ...all the fiber optic cables with software? We aren't going to move everything to the cloud, including the cloud?

    • Re: (Score:2, Funny)

      by Anonymous Coward

      ...all the fiber optic cables with software? We aren't going to move everything to the cloud, including the cloud?

      Better be careful. God's gonna disappear in a puff of logic if you're not.

    • by fuzzyfuzzyfungus ( 1223518 ) on Monday April 29, 2013 @08:51PM (#43586667) Journal

      Certainly not all of them; but I'm pretty sure that the box they are all plugged in to is, pretty much, using a software layer to abstract the ugly details of dumping traffic between them over a really, really, fat internal bus of some flavor.

      And, in many cases, a single fiber is(thanks to software) being sliced up into a bunch of little VLANs to create a logical topology that (while it is ultimately constrained by the physical one) is substantially different than the physical topology, especially once you count aggregated port groups, redundant links, and so on.

      'SDN' doesn't mean jack in part because everything except your 20 year old 10Mb hub is already doing some amount of software trickery(even dumb switches keep track of which MAC(s) are on which port, and anything with 'managed' in the title can do quite a bit more), with varying levels of ASIC vs. general-purpose-CPU and varying levels of correlation between the logical topology and the physical topology.

      There just isn't a nice bright line(at least in terms of real-world use cases, obviously a VM chattering to itself over a loopback interface is 'software' and a passive ethernet tap is 'hardware') between what is 'software defined' and what isn't. They all obviously depend on hardware to execute the software; but the amount of additional logical complexity slopes up surprisingly smoothly.

      • by swalve ( 1980968 )
        I'm guessing the goal is to use more or less comoddity hardware. Why build an ASIC when you can solder some broadcom chips to a backplane and do all the fancy shit in software?
      • by Lennie ( 16154 ) on Tuesday April 30, 2013 @07:05AM (#43588987)

        SDN in practise just means, networking things (private networks, VPNs, loadbalancers, etc.) have an API so they can be automated.

        So when you need to scale out, because your website has more visitors during the day then you don't just get new VMs but those VMs also get connected to the right networks or extra load balancers gets added as well.

        The software in software defined networking, is the application specific software. That application can be that website as mentioned above or something completely different.

        For example Google uses their self-developed software to reserve bandwidth for their different applications and data-replication jobs and handle link failover on the WAN-links between their datacenters.

        Because they used OpenFlow their were able to save money on their WAN-links because they get better utilization than traditional methods. They have normal Google servers that 'directly' configure the forwarding tables.

        • by Shatrat ( 855151 )

          Thanks for actually posting an intelligent comment. Everytime there's a story that involves technology I work with I realize how ignorant most of Slashdot really is.
          The API I think is the key observation. Forget websites though, that's chump change.
          SDN is actually really interesting for my industry, long haul fiber networks. Today we have multiple layers of equipment, the physical fiber plant, the DWDM layer, the OTN layer, the Sonet or Packet transport layer, the IP/MPLS layer. Today those layers don't

          • by Lennie ( 16154 )

            Websites was a an example of 'cloud computing', I wouldn't call Amazon AWS and the others chump change.

    • by istartedi ( 132515 ) on Tuesday April 30, 2013 @12:49AM (#43587579) Journal

      We aren't going to move everything to the cloud, including the cloud?

      Sure we can. It's all based on something I call the "metacircular evaluator". My consultancy and I can install MEs in all your software systems so that you can move the software into the software, and define your business in terms of your business. "My god!" the tech business reporter exclaimed, "this is the most revolutionary thing I've ever heard, tell me more about how these MEs work".

      Well, you just have to re-write everything in Lis--. And then, before I could even finish, the room was empty.

    • There's a yo dawg / turtles all the way down in here somewhere....

  • by gatkinso ( 15975 ) on Monday April 29, 2013 @08:43PM (#43586625)

    ...no reason not to be honest.

    • by girlinatrainingbra ( 2738457 ) on Monday April 29, 2013 @09:26PM (#43586839)
      SDN is the hot new buzzword, just like "cloud" computing has been for the last few years. Buzzwords fly by. I agree with you that he can afford to be honest, but he's not just being honest, he's pointing out the "fuzziness" of what the term "SDN" is being applied to.
      I believe he had a fixed and set definition which he must have specified with some detail in his thesis (isn't that what you're supposed to do in a thesis? be specific?), but nowadays everyone and anyone is calling any configurability of the top or higher levels of networking as "Software defined networking".
      • by Lennie ( 16154 )

        Software defined in my mind just means, it has an API so that application specific software can control it.

        • re: it has an API so that application specific software can control it.

          That is pretty much what I meant by "configurable"! You just expressed it more clearly and specifically, as one ought to in a thesis or in a thesis statement! I strongly agree with you.

    • by AK Marc ( 707885 )
      He was never dishonest. The problem was that people not wanting to sound dumb, don't question buzzwords. I've shut up every SDN proponent I've run across by asking "what does this do that couldn't be previously done with an ACL?" Nobody could give me an answer. I'd be interested if anyone here can tell me one thing SDN does that a layer-7 ACL couldn't do (assume the functionality of an ACL, but with DPI for the trigger case).

      The closest I've come is having a switch sniff for SIP traffic and setting up
      • by Anonymous Coward

        If the CAM (Content Addressable Memory) is wide enough to also match on application data in the packets, you can use switches to deliver network packets to the correct server without using the normal network addresses:
        - for example a load balancing switch (low latency and cut-through (up to the end of the matched data of course)).
        - A network middle ware for pub-sub architecture which can handle a higher number of data-channels.
        * instead of using the multicast addresses one could use partial m

  • by Anonymous Coward

    I hope that link isn't to the actual thesis paper, because it's only a few pages and describes the most obvious thing in the universe, and more importantly nothing that hadn't been written like a 1000 times before. I hope that his little toy network VM described in that article isn't the genesis of "SDN", because that would either make "SDN" complete bull* from day 1, or make him and everyone else look completely idiotic.

    • by tibit ( 1762298 )

      It's not bullshit, it's just packaging of existing technologies, for a particular purpose. An SDN client-server connection, apart from housekeeping, is just TCP/IP encapsulated ethernet traffic. Pretty much like remote TUN/TAP. It was designed to solve a particular problem: that of letting students operate on live internet traffic from their own development environment, without having to muck with low-level platform details of how you get the ethernet packets in and out. Turns out it was useful for a few mo

  • dunno (Score:2, Interesting)

    by Anonymous Coward

    So the only benefit to "SDN" whatever it is that I can tell is that it will could possibly allow source routing. The existing protocols basically will route your packet the shortest hop way or another under guidance of some other metric, unless you set up the router to do some hacks (I hear). The setting up part is done by a human, a network engineer, and the SDN folks think that it shouldn't be done by a network engineer, it should be done by end point software because the network engineer is a human so

    • Re:dunno (Score:4, Funny)

      by postbigbang ( 761081 ) on Monday April 29, 2013 @09:53PM (#43586941)

      You mean "infecting" servers with router VM appliances that smoke government blocks by creating backdoor VPNs, proxies, shadow VLANs and stuff?

      Never happens.

    • Re:dunno (Score:4, Informative)

      by ppanon ( 16583 ) on Tuesday April 30, 2013 @01:05AM (#43587627) Homepage Journal
      From what I read of SDN, the idea is to have centralized routing (presumably for use within a data centre, telco, or high-performance campus network) instead of decentralized routing. Instead of having each individual node recalculate routes using tree-based routing algorithms like OSPF, a central node with a holistic view of the network recalculates and redistributes routes using algorithms that allow more fine grained slicing of packet flows for closer to optimal load balancing and congestion management.Unless you're a telco, a co-lo, or have a datacentre with >5 racks steadily generating >50Gb/s of network I/O per rack and needing high availability, it's doubtful that you need to pay the premium for it.
  • by VortexCortex ( 1117377 ) <VortexCortex&project-retrograde,com> on Monday April 29, 2013 @08:59PM (#43586719)

    I'm glad he's laughing all the way to the bank. Gives me room for my new buzz-word compliant technology: Hardware Optimized for Software Systems (HOSS)

    Shhhh, it's just ASIC in sheep's clothing.

  • If you wanted your new buzzword to have a real meaning perhaps you should have named it something that actually means something. The words Software Defined Network have a generic, non-specific meaning, that's why they are being applied to everything that even remotely fits their definitions. Whatever happened to real names with specifics, like "Carrier sense multiple access with collision detection"

  • by hackus ( 159037 ) on Monday April 29, 2013 @09:44PM (#43586903) Homepage

    From what I can tell it is the idea of having all of the routing centralized at one location with nodes which just accept the commands to route certain src and dst streams. It is different because the software defines the routing on a server in a logical representation for centralized management while the nodes are just really hardware appliances.

    It is a nice idea to reduce cost, but in my opinion this is where you would never want to do something like this because it allows way too much power in a central authority.

    It would be a Chinese government dream network though and the NSA/CIA would piss themselves that ever happened.
    (i.e. In such a system the distributed BGP internet would just go away.)

    I am totally against it, and I think everyone will be after they see what the real intent is: To bring network layer control through software to a central authority, which isn't possible right now, and once done, shut it down whoever isn't in the 1%.


    • Re: (Score:2, Informative)

      by Anonymous Coward

      You have zero idea what you are talking about. OpenFlow is a wipe of 25 years of networking crap. It does not hand over control to the government anymore then running BGP on your network does. The idea is that within your network, lets say you are a datacenter operator, you can impose policy over a group of switches and routers. They become a dumb packet forwarding engine. Why would you want to do this? Lets say you are an operator selling VMs. You offer an up-sell option of providing a FW or SLB to the cus

      • Re: (Score:2, Informative)

        by Anonymous Coward

        congratulations. you've just substantially raised the complexity of the installation and gained absolutely nothing
        in terms of performance, or policy expression, or security.

        have fun masturbating

        • Re: (Score:2, Informative)

          by Anonymous Coward

          oh yes, and drastically reduced the fault tolerance

          • by ppanon ( 16583 ) on Tuesday April 30, 2013 @01:27AM (#43587685) Homepage Journal

            In theory, the centralized routing agent, by having a global view of the network, can both optimize load across links and adjust to congestion patterns better than a distributed network infrastructure can. So for telcos and other colos, that can mean better uptime and more efficient use of expensive high-end infrastructure. Just as a centralized traffic light control system for a city can identify congestion hot spots and adjust light timing to reduce congestion better than isolated traffic lights can.

            For instance, if nets A, B, and C are linked in a loop for redundancy, spikes of traffic between A and C could be load balanced by shifting certain source-destination pairs over the link to B to avoid congestion, even though it would normally be faster to go over the A-C link and that's what an OSPF router would do. Alternatively, in the longer term, a telco could make more effective use of fibre if it could temporarily re-allocate fiber currently not used for voice to deal with a spike in data traffic.

      • The funny thing is that sort of stuff has been done for years in VM hosting without all the virtual network bits. Early last decade hosting providers were using routing protocols to shift inbound traffic about and rewrite tricks for the outbound to make sure bits got to the right physical went though the right firewall or load ballancer or what have you. Well before this kid wrote his thesis.

        As to bigswitch there best hardware I could find was a handful of 40ge ports and a few 100ge on a "core" switch tha

  • by Animats ( 122034 ) on Monday April 29, 2013 @11:27PM (#43587351) Homepage

    OpenFlow is basically a way to turn a packet network into a rather dumb virtual circuit network. It works something like Tymnet, circa 1971. In Tymnet, all the virtual circuits were set up by a "supervisor" computer, which told each node where each flow was supposed to be forwarded. The supervisor also handled authentication, but data packets didn't have to pass through the supervisor once the connection was set up. That's what OpenFlow does, mostly. The first packet of each new "flow" (IP/port/IP/port set, usually) is sent to Master Control, which decides whether that flow will be allowed. Master Control can also choose to monitor the flow. The implications are obvious.

    DOCSIS 3, the cable modem traffic control architecture, can potentially do most of the same things, and offers better control over bandwidth. DOCSIS 3 tends to be run more to control users than to maximize throughput, but that's a marketing issue. (If your cable connection is throttling something, the commands to do it were probably sent to a DOCSIS node.) There's good QoS and fair queuing stuff in DOCSIS 3, but it's not always used intelligently. DOCSIS is less intrusive than OpenFlow; the nodes are sent rules to enforce, but there's no need to get permission of Master Control for every new flow.

    The rest of "software defined networking" seems to involve adding another layer of indirection to Ethernet addresses so they can be moved around within the data center. ("There is no problem in computer science that cannot be solved by adding another layer of indirection.") That's a reasonable network management tool, but it's not exactly a profound concept.

    • by Lennie ( 16154 )

      With OpenFlow you can preconfigure most of the forwarding entries (not just routing) as well.

  • It is used (without knowing they are using it) by people that avoided legacy one router/cable per task Cisco mentality.
    Remember Cisco 1700/2600/3600 and how doing ANYTHING on them cost you ass load of money? It was never a question of what should we do, it was always a question what thingamajiggy we should buy to do this one specific thing we want.
    Never generation stopped to give a shit about Cisco/Agere/Lucent/Juniper and started deploying all manner of embedded Linux/BSD with software routing. Sure it is slower than dedicated silicon, but it is also much cheaper and more flexible.

    • by Lennie ( 16154 )

      SDN isn't about commodity hardware per se, it is more about having an API to configure/control and especially automate the network.

What is algebra, exactly? Is it one of those three-cornered things? -- J.M. Barrie