In my discussions with service providers of all sizes, there is clearly frustration that SDN and NFV are being held up by some as holy-grail solutions for declining per-bit delivery revenue. And while I agree that these technologies are critical to future success, they are merely tools on the drive toward the real holy grail: automation. In this two-part post, I’ll argue that automation is a continuum requiring constant investment and small-step implementation. Waiting for a future enabling technology or clear standards could prove fatal. New tools and standards are certainly necessary and will eventually come, but there isn’t time to wait.
Bandwidth demand has skyrocketed and the cost-per-bit for services delivered is rapidly approaching the revenue-per-bit generated (and soon may exceed it). Carriers have understandably been looking to the cloud for help in achieving new levels of automation and efficiency. The buzz around NFV, SDN, orchestration, and Central Office Re-architected as a Data Center (CORD) is very high. But many industry pundits question if these technologies are truly ready for prime time, and agree with me that carriers must move quickly, with tools currently available, while simultaneously pushing for industry-wide standards.
MSOs long ago learned that to compete with telcos, they had to reduce both operational and capital expenditures. Thanks to little competition between them, MSOs formed CableLabs, an industry-sponsored body that, among other things defined DOCSIS. This standard specifies architectural, data, control, and management plane interfaces, and all but ensured commoditization and optimization of service delivery. DOCSIS not only laid the groundwork for rapid automation but by creating a cookie-cutter approach to network equipment, resulted in direct consumer benefits.
Unfortunately, telcos don’t have an equivalent industry initiative and have somewhat cobbled together standards and recommendations from organizations such as ETSI, ITU, TMF, and BBF. But those lack the detailed specifications needed to optimize automation the way DOCSIS did.
This does not mean that telcos are at a fatal disadvantage when it comes to automation. It simply means that forward progress must be made at the same time the standards are being developed, even if they must be adapted later to fit a newer model. In fact, the move toward SDN and NFV will, by nature, necessitate that certain standards be established, either by the larger players who implement first or by an industry consortium.
Next week, I will outline some ideas for how carriers can begin the drive to automation immediately which will allow high-cost resources to either be eliminated or better, re-deployed to support value-add services and create new revenue opportunities. Until then, please let me know your thoughts on how process automation falls within your priorities.