Telco Cloud-Native: Revolutionizing Telecommunications
The telecommunications industry has undergone profound transformations over the past decade, spurred by the growing demand for higher reliability, agility, and enhanced performance.
The establishment of the ETSI Industry Specification Group (ISG) NFV in 2012 marked the onset of the cloudification journey for telecom networks. After five years of theoretical research and practices, operators began deploying cloudified core networks at scale in 2017, marking a significant milestone in NFV technology. The entry of information technology (IT) and public cloud vendors into the telecom sector, coupled with the convergence of IT and communications technology (CT), has propelled the adoption of cloud-native principles among operators, steering the evolution of telco cloud towards containerization and automation.
This evolution has paved the way for the emergence of Telco Cloud-Native. As we delve into what telco cloud-native entails, its key features, and its focuses, we'll gain a clearer understanding of how it is set to shape the future of telecom.
What is Telco Cloud-Native?
Telco Cloud-Native integrates the principles of cloud-native — initially derived from IT practices — with the unique needs of the telecommunications sector. This innovative approach leverages cloud-native technologies — such as microservices, containers, and declarative APIs — to craft scalable, high-performing, and agile telecom services. While elasticity and agility are paramount, operators also recognize the importance of carrier-grade requirements, end-to-end (E2E) automation, and operational intelligence to fulfill the demands of modern telecommunications.
The foundation
Virtual machines (VMs) have long been the backbone of server virtualization. Each VM runs a complete operating system (OS), along with the necessary applications and libraries. While this provides isolation and compatibility, it also means that VMs tend to be resource-intensive, requiring significant overheads in terms of CPU, memory, and storage.
Containers, on the other hand, encapsulate an application and its dependencies in a lightweight package. Containers share the host OS kernel, making them significantly more efficient. They use less system resources, start up in seconds, and can run in diverse environments. This fundamental difference is what makes containerization so appealing for modern telecom application development.
While operators are rapidly transitioning from Virtual Network Functions (VNFs) to Cloud-Native Network Functions (CNFs), the evolution path towards containerization is not that certain and the journey is fraught with challenges. The container standards landscape is currently fragmented, leading to varying approaches among organizations. Some prefer bare metal containers for new deployments, while others opt for a transition to VM containers. Furthermore, the significant market presence of existing VMs presents a challenge, requiring operators to balance innovation with the need to protect their investments in legacy infrastructure. In this situation, a seamless evolution solution represents the most compelling choice. To address these complexities and facilitate E2E automation and unified management of both VM and container resources, industry associations such as ETSI ISG NFV are actively driving advancements in NFV and container standards.
Key features
2. Converged architecture: Given the widespread adoption of VMs in telco clouds and specific security and storage requirements for certain telecom services, VMs and containers are likely to coexist in telecom networks for the foreseeable future. A telco cloud-native network must efficiently manage both on one cloud.
Huawei's converged dual-engine architecture exemplifies a best practice, facilitating a clear view of global topologies, dynamic scheduling of VMs, containers, and bare metal servers as demanded, along with diversified computing power. According to a report by ABI Research, in a typical 5G Core (5GC) bare metal container project, dual-engine architecture can reduce capital expenditure (CAPEX) by approximately 32% compared to traditional deployment solutions, effectively safeguarding operators' investments.
3. Carrier-grade reliability and performance: Emphasizing carrier-grade reliability, telco cloud-native ensures that services remain accessible even during hardware failures. Innovations in storage management and operational capabilities enhance the overall resilience of telecom networks. Full-offloading architecture with data processing units (DPUs) allows the central processing unit (CPUs) to focus on more compute-intensive tasks, thereby improving overall system performance.
4. E2E automation: After a decade of NFV development, operators have reached a consensus that technical architecture, whether through hierarchical decoupling or public cloud, is only a pathway towards cloudification. The essence of a telco cloud lies in E2E automation, which aims to minimize network faults caused by manual errors, enable rapid fault location and service recovery, and facilitate agile delivery and efficient O&M. This approach allows operators to concentrate on innovation rather than manual intervention.
5. Simplified delivery: A cloud-native architecture should empower operators to dynamically scale services in response to changing demands. Furthermore, it must facilitate the agile rollout of data centers (DCs), allowing operators to introduce new hardware without necessitating upgrades to existing inventory nodes.
6. Intelligence: The integration of artificial intelligence (AI) and intelligent orchestration enhances operations and maintenance (O&M) capabilities, paving the way for comprehensive E2E automation.
Three focuses of cloud-native
As the telecom landscape continues to evolve, so do the objectives and capabilities of telco cloud-native. During this evolution, three key focuses have emerged: Available, Easy to Use, and Intelligent Use.
1. Available: The initial focus centers on achieving carrier-grade reliability and performance on cloudified networks. Historically, NFV faced challenges like limited application isolation, inadequate scalability, and low deployment efficiency. Years of practice have led industry leaders, such as Huawei, to revolutionize reliability and performance, achieving a groundbreaking 99.999% carrier-grade reliability and facilitating the growth of large-scale global telco cloud infrastructure.
Notably, Huawei's storage bypass feature enables the system to run as usual for 4 hours in the event of a storage device fault.
Additionally, a preferred telco cloud-native network should allow operators to visualize network topologies, encompassing computing, storage, and network resources, as well as cloud platform infrastructure, VNFs/CNFs, and the network management layer. It should also facilitate centralized resource management. With an industry-leading dual-engine architecture, Huawei's telco cloud platform is uniquely positioned to accommodate both VMs and containers, deploy different services independently, and enable on-demand OS upgrades — making it the optimal choice for operators looking to streamline their operations.
3. Intelligent Use: The advent of virtualization and cloudification technologies has complicated telecom network deployments. With the rise of the autonomous network (AN) concept, operators are now increasingly considering intelligent O&M services, including intent-driven network function lifecycle management and a unified foundation for AI capabilities.
By utilizing AI and machine learning (ML), an intent-driven network can analyze vast amounts of data to predict network behavior, optimize performance, and proactively address issues to realize closed-loop fault handling. A unified AI capability foundation should accommodate heterogeneous deployments across both x86 and Arm servers, along with diversified computing power facilitated by XPUs (CPUs, DPUs, and GPUs).
In this context, the NFV framework, as outlined by ETSI, can be further simplified with the introduction of unified cloud management (MANO 2.0), which empowers intent-driven networks, the management of heterogeneous hardware, and the seamless orchestration of resources spanning infrastructure to network functions, and from the central sites to the edge. By leveraging AI, telecom providers can elevate network planning, maintenance, and optimization to the next level.
Conclusion
The shift to Telco Cloud-Native is no longer just an option; it's a necessity for telecom operators striving to stay competitive. By concentrating on a unified approach to standards and interoperability — guided by organizations like ETSI — the industry can prevent fragmentation and reduce trial-and-error costs.
The journey toward Telco Cloud-Native is a collaborative endeavor, requiring industry stakeholders to come together to shape a resilient, elastic, and intelligent future for telecommunications. Embracing this evolution, operators can harness the full potential of cloudification and deliver unprecedented value to their customers.
Learn more about Huawei’s Telco Cloud Solution.
Disclaimer: Any views and/or opinions expressed in this post by individual authors or contributors are their personal views and/or opinions and do not necessarily reflect the views and/or opinions of Huawei Technologies.
Leave a Comment