As a passionate telecommunications service provider advocate, my racing colours are unambiguous. I want to take the latest of technology innovations into our multi-tenanted core -- and, within a managed services wrapper -- offer our customers a unique experience, technically, operationally and commercially.
As many would hopefully understand, this is not easy to do, that is, making significant capital investments involved in product development and platform deployment. It's generally better for the service provider if productisation can be embarked upon based on one, maybe two maximum technology options. Even with two options, hopefully over time (remember IPVPN RFC2768 vs RFC2547bis?), one should emerge victorious, allowing the SP to focus their efforts uniformly on one approach.
As per my previous blog, we are evolving into an age where customers are able to procure products/services a different way. They can also now procure the lego blocks that underpin them. We are in the age of the virtualised function/application marketplace, some would argue about to drown in options (do you really need six vRouters?) but ultimately we will be providing unprecedented customer agility, utility and choice.
This can of course get confusing on a number of levels. And after sitting through seven different vendor SD-WAN presentations on a recent US trip and seeing a number of others on trade show floors etc, a few things become clear to me.
As a service provider we need a way to have a few more jockeys on some of these SD-WAN horses. The winners are not clearly emerging from the pack -- just yet. To this end, we are presently supporting at last count, five different SD-WAN vendor solutions via our advanced technology practices arm BTS. Furthermore we are looking to productise perhaps two of these solutions within unique SP offerings, this coming financial year.
For service providers, complexity may extend beyond the number of functions in the marketplace, that they need to be support orchestration onto varied core/edge and premises networking infrastructures. They may also need to base offerings on a combination of open and proprietary underlying platforms to get the best end customer outcomes.
In May I discussed the coming together of PEN (OpenDL/OpenStack/OpenFlow) and Symphony (Cisco VMS/Tail-f) within the Programmable Network initiative. My view coming back from the US, is that the latter could indeed remain the best option for "Cisco on Cisco" function/platform orchestration (e.g. Cisco iWAN).
The new age is posing some real questions around SP managed services. We are not only in the VNF age becoming a software provider, we also need to be a provider of advisory services to help our customers get the most from these new applications.
It's indeed hard to not be super impressed with drag and drop service model construction and "slider based" application visibility and control. However I think our customers may just want/need a little assistance getting the most out of the new age technology stacks.
When I was playing with them, I certainly needed some help and we have found in the past, with technologies like WAN Optimisation, early portal/tool excitement can evolve into almost network solution neglect as customers get in some cases just too busy to tune/optimise their networks as application profiles change.
I hope you are looking forward to coming with us on the SD-WAN journey. It is one of seven priorities within our Programmable Network initiative where we will develop capabilities aligned to five key principles. I look forward to discussing these principles and priorities in future blogs.
For more networking, visit Telstra Exchange.