Are Your Systems Truly Cloud Native? There Are Ways of Knowing! – Part 2/3

In Part 1 of this three-part series, we examined how the telco journey toward cloud nativeness is critical in unlocking the transformative potential of 5G. While many telcos are on this path, the challenge lies in identifying and implementing truly cloud native solutions. In Part 2, we focus on what it means for a software system or function to be cloud native, the pitfalls of repackaging legacy applications and the specific challenges telcos face in evaluating their vendors’ offerings.

What Does It Mean to Be Cloud Native?

“Cloud native” is more than a buzzword; it represents a fundamental shift in how software is designed, deployed and managed. A cloud native software system or function typically exhibits the following attributes:

• Loosely Coupled Microservices Architecture: Systems are broken into small, independent components that communicate via APIs, making them easier to scale and maintain while also providing fault isolation, preventing cascading failures.
• Declarative APIs: These APIs facilitate seamless communication between services and ensure compatibility with modern orchestration platforms.
• Packaged as Containers: Containers provide a consistent environment for applications, ensuring they run predictably across different computing environments.
• Orchestrated by Platforms Like Kubernetes: Orchestration ensures scalability, reliability and resource optimization by automating the deployment and management of containers.
• Immutable and Elastic Infrastructure: Cloud native systems run on infrastructure that is easily replaceable and dynamically scalable to meet demand. Following the “cattle, not pets” philosophy—resources are treated as disposable and replaceable rather than uniquely configured. This approach ensures that infrastructure changes are managed through automation, enabling resilience, consistency and rapid scaling to accommodate varying workloads.
• Highly Observable: Systems are built to monitor performance, diagnose issues and adapt in real-time. Observability typically includes logs, metrics and distributed tracing to ensure deep system visibility.
• Resilience and Fault Tolerance: Cloud native systems are designed to anticipate, tolerate and recover from failures without impacting the overall functionality. This includes mechanisms like self-healing, retry logic, circuit breakers and distributed systems design that ensure high availability and minimal downtime, even during unexpected disruptions.
• Scalability: Cloud native systems are designed to scale both horizontally (by adding instances) and vertically (by increasing resources) to handle varying workloads effectively. This ensures optimal performance and cost efficiency, even during traffic spikes or growth. Scalability is often achieved through distributed architectures, stateless components and automation.
• Leverage Automation and DevOps: Continuous Integration/Continuous Delivery (CI/CD) pipelines and DevOps practices enable rapid iteration, deployment and improvement of applications.

These attributes work together to deliver the scalability, agility and efficiency that telcos need to stay competitive. However, being cloud native is not a one-and-done achievement. It is a dynamic state, especially within telecom networks where these paradigms are being introduced for the first time. Telcos must align themselves with industry standards and participate in initiatives like the Linux Foundation Cloud Native Telecom Initiative (CNTi) to remain at the forefront of innovation.

You Can’t Just Repackage Legacy Applications

For telcos evaluating vendor offerings, the challenge is clear: How can you differentiate between a truly cloud native solution and one that has simply been retrofitted to check the box? The good news is that there are vendor-neutral technology initiatives that provide guidance on cloud native best practices, and these are definitely worthy of consideration to speed the process and deliver improved outcomes.

What Does It Take to Make Telecom Networks Cloud Native?

It goes without saying that telecom networks operate at a different level than the typical web application or service. Here are some of the factors that must be taken into account when implementing cloud native network applications.

• Performance Requirements: Telecom functions typically involve real-time data processing that requires extremely low latency, high throughput and deterministic, predictable performance.
• Stateful Workloads: Telecom functions like session management and user mobility are inherently stateful, making it challenging to scale and orchestrate. In addition, state consistency across distributed instances must happen in real-time, which is resource-intensive.
• Scalability Models: Telecom systems have historically scaled vertically by adding resources to a single instance, while cloud native approaches for web applications scale horizontally, spinning up multiple instances. This paradigm adjustment for network functions can be tricky. Further, network traffic is highly dynamic and unpredictable, demanding more robust and responsive scaling mechanisms.
• Resource Optimization: Telecom workloads often require specialized hardware like FPGAs, SmartNICs or GPUs to meet performance demands. Integrating these into cloud native environments is more complex than general-purpose computing.
• Network Complexity: Advanced telecom services rely on features like deep packet inspection, network slicing, service chaining and network overlays, which are far more complex and require significant management overhead than the simple request-response interactions in most web applications.
• Reliability and Availability: Telecom networks typically promise five nines availability, translating to only a few minutes of downtime per year. As such, they must quickly recover from failures to avoid disruptions in critical communications, which adds layers of complexity.
• Standardization and Interoperability: Telecom networks have always included, and will for years to come, legacy hardware and software that need to interoperate with new cloud native components. With diverse and fragmented vendor solutions, there will long be a need to manage that fragmentation while adopting unified cloud native practices.
• Security Requirements: Ensuring strict tenant isolation in a multi-tenant environment is vital for telecom networks to protect sensitive user data and comply with stringent regulations (e.g., lawful interception, data localization) that go beyond the usual compliance requirements for most web applications.
• Development and Deployment Lifecycle: Implementing continuous integration and delivery (CI/CD) for telecom CNFs is complex due to stringent high availability testing and validation requirements. In addition, cloud native tools like Kubernetes were originally designed for stateless web applications and may need significant adaptation to handle telecom-grade workloads.
• Cost and Investment: Transitioning to cloud native requires significant investment to re-architect network functions and retrain teams, which can require additional OpEx that may not be readily available in challenging economic conditions.

Vendor-Neutral Evaluation Tools Can Help

Wouldn’t it be helpful to have a set of open-source tools and processes to objectively evaluate a CNF’s cloud native capabilities? Enter the Linux Foundation’s CNTi which is working to establish vendor-neutral frameworks to address these challenges. These tools will allow telcos to systematically assess critical aspects like container orchestration, networking and performance, ensuring their investments deliver the promised benefits.

Stay tuned for Part 3 of this series, where we’ll explore CNTi’s work in detail. We’ll discuss how their tools and frameworks are shaping the telecom industry’s transition to cloud native, making it easier for telcos and their vendors to improve the cloud nativeness of their solutions and fully realize the potential of 5G Core.

The journey toward cloud nativeness isn’t just about technology—it’s about collaboration, trust and innovation. Telcos and their CNF vendors will succeed together by embracing open-source solutions to future-proof cloud native networks using common language and frameworks that are mutually understood. Those languages and frameworks are available now, so now is the time to act.

Read more about MATRIXX Software’s monetization solution.

---

Olivier Smith
As a member of the CTO Office at MATRIXX Software, Olivier is actively engaged in open-source communities in the Linux Foundation, where he helps advance and promote cloud native best practices for the telecom industry. He is a founding member of the Cloud Native Telecom Initiative (CNTi) and a Technical Steering Committee member.

About MATRIXX Software’s CTO Office
MATRIXX Software delivers a modern converged charging and digital monetization solution proven at scale. Global operators like Telefónica and Telstra, IoT providers like Tata Communications and network-as-a-service providers like DISH rely on the platform to overcome the limitations of traditional Business Support Systems. MATRIXX actively contributes to advancing industry standards and best practices for the next era of telecommunications through our leadership and engagement across 3GPP, TM Forum and the Linux Foundation.