Lift-and-shift: Why does it undermine BSS-on-cloud transformation?

5G opens up vast revenue opportunities for communication service providers (CSPs). However, taking advantage of the large number of potential 5G use cases can introduce complexities that are hard to manage unless CSPs are using business support systems (BSS) that are truly cloud-native. Several CSPs have already started moving their BSS/OSS workloads to the cloud. The ability to rapidly launch bundled digital lifestyle products, create new services offerings and update product pricing, packaging and promotions in real-time combined with the continuously updated software gives them unprecedented time-to-market advantage against their competition. A cloud-based BSS system also enables them to scale up and down rapidly based on demand, making it possible to offer more cost-effective solutions and excellent customer experience. But when it comes to migrating to the cloud, many software vendors and CSPs in the past decade have taken a 'lift-and-shift' approach, which involves taking modular applications and containerizing them in the belief that they are cloud-ready. That sounds like a great tactic -- until it doesn't work. A 'lift-and-shift' approach is bound to fail for many reasons. The result is that CSPs do not realize any of the CAPEX or OPEX advantages they expected from moving to the cloud. Why lift-and-shift doesn't work In a lift-and-shift approach, several critical elements are required to make a CSP’s migration successful. The first and most important ones are security and privacy. A lack of understanding of data protection laws and how they will apply to cloud-based BSS implementations in different geographies can have massive legal repercussions. Non-compliance with data protection regulations, like the EU’s General Data Protection Regulation (GDPR) when handling personally identifiable information (PII) of people, can lead to huge penalties. It is crucial for CSPs to understand what they can and cannot do with customer data when putting BSS on the cloud. CSPs also need to consider performance challenges. For example, not all on premise applications will perform well on the cloud, including policy and charging rules function (PCRF) and network provisioning. It is therefore vital to assess application performance before taking the plunge. That's not all. Cost escalation is one of the reasons why the 'lift-and-shift' approaches fail. Several factors, including skills gap, miscommunication and lack of training, can lead to cost escalation. Unless there is a proper plan in place, costs can escalate rapidly on the cloud, removing any promise of cost efficiency. Further, a lack of expertise is another critical area that needs to be addressed before making the shift to the cloud. While CSPs typically have strong expertise and capabilities when it comes to traditional on premise BSS solutions, making the shift to the cloud without retraining or seeking external help is a recipe for disaster. What does it take to make it work? So, how can CSP make a success of moving their BSS to the cloud? A comprehensive and systematic approach is required to identify the right platform and ensure a smooth BSS migration, based on a well thought out design-construct.

1. Technical Evaluation - Application Maturity Not all applications may be suited for the cloud and its multi-tenant architecture. Therefore, the first step of the journey is to do a technical evaluation of the application’s maturity and whether it has the functional and technical capabilities for cloud readiness. The deployment stack should be feature-rich and enable consistent user experience across configuration, provisioning, operations, administration and management. Furthermore, the deployed solution should adhere to zero-code principles to completely eliminate customization. This brings additional benefits of a zero-touch rolling update, enhanced security and cost optimization while ensuring that the latest version runs in the cloud at the same time.

2. Business & Operational Evaluation Once an application has been tested for technical suitability, it is essential to do a business and operational evaluation by conducting a thorough multi-tenant vs multi-stack analysis. CSPs present in multiple countries/regions, or managing different solutions for their retail and brand-challenger customer segments can look at namespace based deployment models to achieve significant operational efficiency. It is important to evaluate local laws and whether it is even feasible to move applications to the cloud for legal reasons such as data localization. A hybrid approach becomes important here wherein the telecom operator could look at moving a part of their technology stack such as billing, CRM and digital engagement platform to the public cloud while core applications such as network provisioning, Policy and Charging Rules Functions and charging could reside on a private cloud native data centre. 3. Evaluate Cloud Providers The next important aspect is to identify the right cloud provider for the application. While the applications built using Cloud Native Computing Foundation (CNCF) based best practices can be deployed in any hyperscale environment, the platform partner must be chosen based on presence, ecosystem and availability of open-source technology components in their environment. To choose the right public cloud vendor, CSPs need to ensure that all the platform components (aside from application micro services and containers) are in place, they get the right pricing, the vendor has a strong regional presence and investment plans while supporting ecosystem applications as well.

4. Cost vs Benefit Efficiency The last but most important aspect of identifying the right platform is cost versus efficiency benefit analysis. CSPs need to chart out their project implementation and long-term operational cost. The ability to weigh up the cost-benefit versus restrictions while running an integrated multi-geography operation is critical. Cloud migration should exploit some of the benefits that platform providers bring (example a data lake, ingestion, a load balancer, registry, service mesh, policy and security) while creating a single glass pane of observability across the workload deployed in multi-cloud environments. Once CSPs have all the pieces together, it is vital to ensure their chosen cloud platform has built-in automation. With 5G services, automation plays a big role in ensuring the network's observability, transparency, and efficiency. The public and private cloud providers also must provide a single glass plane, queue and cache management among other things. It is evident that BSS-on-cloud transformation cannot be undertaken lightly, or without a proper design construct. CSPs need to identify the applications in their environment best suited for the cloud, evaluate the right cloud for each of their applications and ensure that other elements such as privacy and cost are well accounted for. Failure to do so is likely to hamper their cloud migration efforts.