Hybrid cloud platform approach is a key enabler for enterprise reinvention towards business acceleration for Communication Service Providers

Share this post:

Most enterprises are yet to unlock the full value of enterprise transformation, limited to only “greenfield implementation” or “lift and shift” paradigms. There is an urgent need for Telcos to imbibe the technology dynamics driving growth and modernise or replatform monolithic legacy applications that have huge technical debt.

They key drivers for any Communication Services Provider (CSP) to become Digital Service Provider (DSP) can be summarized as below:

a) Digital First Engagement

b) Enhanced Customer Experience

c) Simplified Operating Model

d) Product Supremacy

 For Communication Service Providers, the key set of applications which will drive the above agenda to become Digital service Provider are mainly from the Order to Cash process area.  We have seen that many of these monolithic legacy applications were written around 10-15 years back and carry significant technical debt. Often, the challenges with legacy application modernization typically boil down to cost and complexity. The key impediments to implement above drivers are CSP’s current legacy applications’ inability to address the flexibility, scalability and integration with partner ecosystems.

The way forward is indeed to modernize or replace these legacy applications to become cloud native or use SaaS however it is easier said than done. Over the multitude of engagements that we have experienced in these areas, we have observed the key inhibitors as:

• Inability to update old monolith architectures
• Reengineering legacy applications
• Standardisation

Legacy monolithic apps are difficult to update for architectural reasons. Because all of application’s components ship together, it is difficult and costly to add features given the overhead of complexity and integration challenges resulting in high Opex.

We also observe that there is a tendency to trivialize the reengineering job. While there are tools in play to extract process flows, information models from the source code, extracting business rules poses an entirely different challenge. There is limited capability available with the various tooling options for automated extraction of business rules. Needless to say, without the business rules extraction, the other alternatives would be to engage domain expert and develop the rules from scratch. And that would make whole modernisation lengthy and costly.

Also more often than not, there is little documentation available for these legacy applications nor does  the client IT organization have enough SMEs to validate and confirm extracted business rules. Below is an example to quantify the size and complexity of the problem.

An order orchestration application for fixed line products having client server architecture using java and RMI can have anywhere between 4000 to 6000+ java files and having 750K to 1200K SLOC. Excluding POJOs, DALs and adapter classes, there would still be a need to analyse ~450K+ SLOC to extract business rules. Now from complexity perspective, just for one product (e.g. “wholesale”), there could be 10+ possible status across 20+ steps. With this kind of size and complexity, business rules extraction would be enormously complex and risky. And even if the business rules extracted, two key questions still be to be answered:

• How do we know that rules extracted are exhaustive and complete?
• How to ensure the veracity of the extracted rules?

While business rules extraction is the major challenge, we also cannot ignore the lack of standardisation for process and information models. In the cases where legacy applications were custom built, there would always be a variability across products in terms of:

• Process and step definition (e.g. “Order Capture for Wholesale”, “Order Intake for VoIP”)
• State definition and nomenclature (e.g. “Product Decomposition post validation” vs “Order Decomposition post credit validation”)
• Entity and attribute names (e.g. “Resource” vs “CPE”)

While calling out these above challenges are cliché, however, solution to these problems have not been solved in entirety. In IBM, we have developed a set of offerings merging our research, industry and hybrid cloud capability: –

• Automated Extraction of Bounded contexts and Business entities
• Address the “Veracity” and standardization problems using Telecom Industry reference models
• Loosely coupled, microservices based architecture. Components can be deployed & scaled independently
• Use of open APIs following TMForum standards to model the target process flow
• Bring acceleration in implementing target state architecture using hybrid cloud accelerators.

Our aim towards maximising the benefits of legacy application modernization can be summarized as – improving the velocity of new feature delivery, exposing the functionality of existing applications to be consumed via API by other services, and re-platforming applications from on-premises to cloud for the purpose of application scale and performance in addition to long-term datacentre and IT strategy.

This offering will bring 30% reduction in time to value for Communication Service Providers in their journey towards Digital Service Provider.  We believe this approach is where much of the core value in legacy application modernization can be unearthed.

Telecommunications solutions

Application Modernization

Connect with Chinmohan Biswas LinkedIn

Connect with Subrata Ghosh LinkedIn