Softwarization and the Future of Telcos in Developing Markets

Estimated Read Time: 4 min 

It has been a while since the telecommunication experts came up with the idea of Softwarization, which can be considered as a broader term to describe the concept of virtualization. Surely, the Softwarization is the next major transformation of all Telcos around the world, no matter how big or small they are and the country they operate, it will be a matter of when to make the move.

Please check the below Knowledge Spark on Telco Softwarization for more information.

 

Virtualization is no more an “infant” technology as it is well matured not only as a concept but also as a solution, even though still there are several related aspects which need further improvements. Almost all the major telecommunication equipment manufacturers have shown clear signs of focusing on this concept in their future developments and some have already taken open approaches and come up with their own solutions.

Is it the right time to adopt this concept? Are we lagging behind? Are we going to embrace it ahead of others or wait till someone else tries it out? These are some questions which the Telcos may have in their mind-set at the moment. The common answer for all these questions is “It depends on where you are and what you do right now”.
 
If we narrow down our focus and concentrate more on developing markets in the Asia-Pacific (APAC), they were among the first movers in adopting 3rd and 4th generation mobile technologies even though there was a significant confusion among the Telcos on emerging technologies. Market saturation, level of competition, customer expectations and past experience were among the factors which drove them from behind. In most of the cases, there were happy endings, because being first to the market helped them to harvest in a blue ocean which let them to attract the majority of the newly created markets.

Will this strategy work in Softwarization as well? May be “Yes”. Then why don’t they get it implemented as fast as before? That is the next question most of the people would ask. Well, that also has an answer.

Firstly, it should be realized that Softwarization will add more value from the operator’s perspective, such as flexibility, customization, short development cycles, cost effectiveness and so on. But from the customer’s perspective it adds little or nothing since the customer cares only on the service he/she gets and how it is delivered is beyond their concern.

Secondly, most of the Telcos have already spent a pile of money on legacy systems and hardware, which are yet to be fully depreciated and even the cost has not yet recovered. They feel like it is still too early to throw away the special purpose hardware for some general purpose hardware. That makes it harder to start the journey with virtualization and they may stick to these systems for another few years.

Next, it seems there is a degree of uncertainty among the operators about the new concepts and how long those will remain valid. As described by the Moore’s Law where the power of computing gets doubled in every eighteen months, new concepts and technologies emerge and disappear in short periods. What we see as the best might get out-dated in few years’ time.

Finally, in some cases, the bitter experience of operators may also slow down their speed of adopting new technologies and concepts. I have met several operators who had the bad experience of being the first movers and then becoming a test bed for some equipment manufacturers. These concerns will reduce the attractiveness and the speed of adoption to some extent.

My belief is that even though all these concerns are valid, Softwarization is the way to move forward form where we are right now. The vendor support on legacy hardware reduces day by day while the risk of keeping those increases considerably. On the other hand Telcos are nowadays seeking for more cost effective, less complicated and flexible solutions which gradually drive them towards the virtualization. It can be either tomorrow or some years later, slowly but surely the time will come for them to go for it.

Rajiv Weragama, ARCTube | @RajivWeragama

Note: The 1^st version of this article appeared at  https://www.linkedin.com/pulse/softwarization-future-telcos-developing-markets-rajiv-weragama/

Can we fully "automate" a telecom network?

Estimated Read Time: 3 min

Traditionally, the Operation, Administration and Maintenance (OAM) of telecom networks and systems have been hugely dependent on humans. Human intervention is required at different levels, mainly at hardware and software. The dependency is more prevalent with wire-line operators as compared to wireless operators. While this had its benefits in the past, with multiple changes of different magnitude happening in the surrounding, too much dependency on humans on telco operations has sometimes become really inefficient and unproductive. 

Telecom operators use different business models to source the human services required. These include in-house, partly outsourced and fully outsourced, etc. But in the long run, all these models have their own defects due to various reasons. As telecom networks transform from Circuit Switched based (mainly TDM) OGNs (Old Generation Networks) to Packet Switched (mainly IP/Ethernet) based NGNs (Next Generation Networks) to virtualized NGNs (vNGN) or Software Generation Networks (SGNs), certain amount of the work done by the humans could be automated, increasing the efficiency and productivity of the overall operations.

Please check the below Knowledge Spark on Telco Digital Transformation by ARC Tube.

SGN, which includes virtualization of the network [Software Defined Networking (SDN)] and network functions [Network Functions Virtualization (NFV)], bring in automation and agility to telecom networks at different levels. With proper orchestration of different resources, an end to end service across different network segments and systems could be provided almost automatically.

What if the user requests a "telecom" service online through a web portal and is automatically provisioned and delivered within seconds?

While online service requests and automatic service delivery is common with Internet based web scale companies, telecom service delivery has traditionally includes many human interventions at different points. This is common for service management as well. There are obvious areas in a telecom domain, where we can NOT practically introduce automation. These include, but not limited to, wires (copper, cable, fiber), towers, antennas, chassis, cards and power supplies. However, once these basic building blocks are in place and properly connected, the rest can be easily automated with proper implementation and usage of software. Software is a very powerful tool and the code has really become the king, not only in telecom space, but also in many other areas.

If the network intelligence is logically centralized, few experts can manage everything from anywhere. Physical intervention is required only when there's something to do with hardware and you can do it with a properly trained set of staff easily. To achieve this, many things need to put in place as listed below.

• Virtualizing networks, network elements and systems wherever possible.
• Abstracting different levels of the network to a software layer using SDN.
• Virtualizing network fictions of the operator using NFV.
• Virualizing the end customer services delivers across the network and systems using NFV and Virtual Customer Premises Equipment (vCPE), where the Virtual Network Functions (VNFs) are located inside the vCPE, at the network edge or in the cloud. The vCPEs are essentially plug-and-play and self-serviced by the customers themselves.
• Orchestrating the resources and the services.
• Re-architecture of traditional Central Offices (COs) or Local Exchanges (LEs) to be mini Data Centers (DC) to host VNFs. CORD (Central Office Re-architected as a Datacenter) can be used for this purpose effectively.

Above need not happen at once or in the order above, but a proper strategy and a plan with realistic timelines can deliver results step-by-step as different capabilities are released at appropriate times.

Anuradha Udunuwara, ARC Tube | @AnuradhaU

Note: The 1^st version of this article appeared at https://www.linkedin.com/pulse/can-we-fully-automate-telecom-network-anuradha-udunuwara/