Although details may vary from platform to platform, SDN essentially involves abstracting the network control plane from underlying hardware -- a truer form of network virtualization than many of the technologies that have adopted that term in the recent past. The practical benefits here are twofold: it makes the control plane remotely accessible, providing greater flexibility to pool network resources for virtual and cloud environments, and it opens the door to third-party solutions in previously proprietary network architectures.
As expected, then, SDN has emerged as a major initiative for smaller networking firms, which are looking to tap into the entrenched holdings that Cisco, Brocade and other giants have built up over the years. Big Switch Networks, for example, is the leading proponent of the OpenFlow standard, supporting the OpenFlowHub development community and recently launching the OFTest framework to gauge compliance among switches and related equipment to ensure optimal interoperability. After all, OpenFlow wouldn't be the first open solution to run out of steam because it failed to meet the performance levels of fully-integrated proprietary platforms.
Of course, SDN has gotten crucial support from large vendors like EMC and VMware, who only stand to benefit if the cost and complexity of linking their primary product lines were to diminish. And leading networking firms appear to be on board as well. Brocade, for example, will support OpenFlow on its MLX 100 Gbps routers and the NetIron OS 5.4 release, while Cisco has said it will include OpenFlow in its open programmable environment for networking (OPEN) platform. Meanwhile, HP is working with F5 Networks to develop "OpenFlow-enabled" platforms.
Of course, the enterprise industry has been down this road before. An open platform gains momentum leading the top proprietary vendors to announce their "support," which in the end turns out to be either lip-service or some form of open/proprietary hybrid that only serves to muddy the waters as to whose version is the real open solution.
SDN could break that mold, however, considering network flexibility is emerging as a key requirement for the cloud.
Already, the technology is finding its way into applications that extend beyond raw networking. Vello Systems, for example, is using SDN as a platform for its Remote Data Replication Manager, essentially creating OpenFlow-enabled network extensions to OpenStack-based clouds. In this way, Vello provides multi-tenant, multi-site clouds capable of self-service provisioning of compute, storage and data protection services.
So is all of this activity an indication that SDN is real, or is this merely the first blush for a technology that looks good on paper but has yet to show its limitations in the real world? The cold hard truth is that it's too early to tell. Open technology lives and dies by the level of support it receives from both the vendor and user communities, so in that regard SDN is off to a good start.
The real test, though, will come when enterprises face the prospect of entrusting the technology with their mission-critical applications.
Arthur Cole covers networking and the data center for IT Business Edge. He has served as editor of numerous publications covering everything from audio/video production and distribution, multimedia and the Internet to video gaming.Originally published on www.enterprisenetworkingplanet.com.