Realizing the Great Promise of Software-Defined Networks

Carl Weinschenk
Slide Show

Interest in Software-Defined Networking Rising Sharply

Like most big ideas, the simple explanation of software-defined networks (SDNs) is compelling.

In current networks, the intelligence that controls where data is sent and the gear that actually sends it are intermingled. This is far less practical than dividing the two tasks and putting the control element at a higher logical point in the network, such as in the cloud. It’s like relying on a traffic helicopter instead of a GPS unit. This enables better decisions to be made – since the higher vantage point allows the network to have a more comprehensive idea of what is happening.

A second advantage is that removing the complexity from each device and basing the commands on interoperable standards means that vendor lock-in will fade.

Simple idea, but extraordinarily complex stuff. One of the organizations aimed at setting the myriad details and establishing protocols of SDNs is The OpenDaylight Project, which includes heavy hitters such as Cisco, IBM and the Linux Foundation. eWeek reports that the group will release details of Hydrogen, its first crack at creating protocols, on Wednesday in New Orleans.

Vendors recognize that future of network architecture hangs in the balance and are acting accordingly:

OpenDaylight is among a number of open efforts around SDN, with others including the Open Networking Foundation (ONF) and the Open Compute Project. Skeptics of the OpenDaylight Project have pointed to these other efforts as part of their criticism. SDN startup Big Switch Networks withdrew from OpenDaylight two months after it was announced, after the project decided to merge the SDN controller technologies from Big Switch and Cisco to create a common controller. Big Switch officials viewed that as an example that OpenDaylight would be driven by big vendors and not by what is best for the industry.

The absence of standards is not stopping the players from getting into the market. This week, Juniper Networks announced a production-ready platform called Juniper Networks Contrail. The very long press release spells out the details of the platform. Most of the release more or less tracks the supposed advantages that SDNs offer:

Contrail is a production-ready SDN solution shipping today that is based on stable and proven networking standards. Contrail creates a virtual network, enabling seamless integration between physical and virtual networks while providing service providers and enterprises with a solution that is simple, open and agile. This network virtualization and intelligence solution has been in trials with more than 40 global customers.

It is important to follow the money when following technology. CIO’s Jeff Vance points out that four big companies have purchased four small SDN startups (Nicira/VMware; Vyatta/Brocade; Cariden/Cisco and Contrail/Juniper). IBM and HP have started their own projects. Vance also suggests who nine of the hottest new companies are, and to boot, mentions the OpenDayLight Project. The nine are Affirmed Networks, Big Switch Networks, Embrane, Jeda Networks, Metacloud, Midokura, Pertino, Plexxi and PLUMgrid.

Add Comment      Leave a comment on this blog post
Nov 24, 2013 11:15 AM Eamon Walsh Eamon Walsh  says:
It is understood that server and storage virtualization have become the norm. We know that virtualization significantly increases system utilization, reduces time for deployment, provides fault tolerance to systems, and has empowered server administrators to manage many more servers than they have in the past by offering a management model that is operationally simple and effective. There has also been direct impact on business continuity planning through the use of virtualization technologies by offering techniques to perform workload mobility. Similarly, I would argue that SDN is necessary for network virtualization, providing the visibility and control of physical network resources needed for the orchestration system to be effective in delivering the virtual network abstraction. The following paper makes the argument that implementing network virtualization by encapsulating the physical network using and overlay mismanages physical resources, limits scaleability, and doesn’t provide effective performance isolation between virtual networks. Reply

Post a comment





(Maximum characters: 1200). You have 1200 characters left.



Subscribe to our Newsletters

Sign up now and get the best business technology insights direct to your inbox.