Telcos across the Asia-Pacific region are adopting technologies like 5G, edge computing, artificial intelligence (AI) and machine learning (ML). Indeed, the pace of 5G network deployments across the region has been striking, with early movers like South Korea pushing the envelope, while China has added massive scale and capacity to cater to new use cases and business models. India and markets in Southeast Asia have also recently launched commercial 5G services, boosting prospects for 5G adoption across the region.

5G adoption has required telcos to implement massive changes in their networks and related infrastructure. This is apparent in multiple areas. At a macro level, the core architecture has moved away from centralized to a distributed architecture, with compute and storage resources moving closer to the end users with edge computing.

The radio access network (RAN) is a great example of this shift, with the disaggregation of the base station pushing new designs with the radio units being decoupled from the baseband. Changes like these also necessitate massive upgrades to the underlying transport networks to ensure seamless movement of data traffic across the network. The core network is also seeing massive changes with telcos implementing cloud-native architectures to increase agility and automation in core operations.

Network automation is a key imperative

With all of the changes being implemented by telcos that are deploying 5G, one of the primary areas of focus for telcos is in the domain of network automation. In many ways, the vision of a near autonomous, light-touch (lights-out!) network that self organizes and heals has been around for some time. However, it is only in recent years that the advent of technologies like AI/ML have pushed the envelope in this area. Many telcos are deploying advanced analytics in their core networks to help them simplify network operations, predict and reduce networks faults and other use cases.

In the Asia-Pacific region, several telcos have invested in network automation. In India, Bharti Airtel is a good example of this trend. India is a hyper competitive market, where demand is very elastic based on the smallest changes in pricing or customer experience. In such an environment, improving the customer experience is critical to maintaining market share, especially for a telco like Airtel which has a legacy of heterogeneous networks (2G, 4G and now 5G), software and tools as well as field personnel that need to be upskilled. 

Identifying network automation as a key imperative, Airtel has set about implementing a new Operations Engine from Ericsson to effect improvements in service uptime and service fulfilment metrics. Airtel claimed 69% of its alarms are now fully automated and resolved, reductions in mean time-to-repair (MTTR) by nearly a third and improved network uptime by nearly half. 

Modern data architectures are crucial for telcos 

Cost competitiveness and zero-touch operations are very important for telcos as they deploy 5G, but this is not the full picture. Monetization is not truly possible until the massive surges in data through the network are stored, processed and analyzed to generate real business insights. This will require fundamental changes in the data architecture deployed by telcos today, with three primary architectural elements that can be implemented.

First, most telcos today either have multiple data repositories stored across their network, or in the case of larger multinational telco groups, they have data repositories spread across operating companies. The former is complicated enough, but the latter can get messy. To start to bring some order to these disparate data repositories, telcos are increasingly looking at the ā€œdata lakehouse.ā€

The lakehouse, as the name would suggest, combines the best aspects and features of traditional data lakes and data warehouses. Moreover, the data lakehouse brings additional features and toolsets like AI/ML, analytics and visualization tools and more. 

Second, telco data users will increasingly need access to data stored in multiple repositories, which in turn could be stored on-premises, in a private cloud or even in the public cloud, depending on the specific application. Accessing this data can be very challenging given variances in data formats and available tools across different clouds. A ā€œdata fabricā€, when implemented well, will connect all the data sources with the data platform, regardless of their location.

Additionally, the data fabric would, offer data users consistent interfaces that are secure and compliant with company governance policies. This ā€œorchestrationā€ of data assets will help data users access data faster, with minimal intervention and tweaking, and process and analyze the data faster. 

Third, while the data fabric can stitch together data assets that are spread across multiple clouds, not every data user is a data science expert or engineer. Data from multiple layers of the stack are sometimes required for business queries within the telco, or even for use by third party enterprise clients and partners. A ā€œdata meshā€ can connect all data assets within a data lakehouse or in multiple locations accessed through the data fabric, and wrap these with advanced tools and features that will elevate the data to business level insights. 

These elements of a modern data architecture will help telcos (re)organize their data but also add orchestration layers that can elevate the data to strategic assets.

A good example of the kind of efforts being made by Asian telcos is the recent work of Telkomsel in Indonesia. Telkomsel is Indonesiaā€™s largest mobile operator with more than 170 million subscribers. The telco was faced with exploding data traffic across its network and surging data from over two hundred data feeds/sources. Faced with data generation of over 50 TB per day, Telkomsel began making investments, in partnership with Cloudera and Wipro, to migrate to a modern data architecture to cope with the increasing demands. The new architecture houses all Telkomsel internal data in a new data lake but also uses Apache NiFi as the data fabric to ingest first, second and even third party data. The net result of this phase of investment has been significant reductions (nearly 60%) in infrastructure and solution delivery costs, as well as improved customer experience and faster app development cycles. All of these improvements are helping Telkomsel better monetize their network investments.Ā 

Original article can be seen at: