Today, Red Hat announced the creation of a cloud-based onboarding service and testbed for network functions, supporting both virtualized and containerized network functions. This effort, in collaboration with Intel, is intended to reflect realistic networking scenarios and provide a true representation of a standards-based (high volume server hardware and open source software) and replicable platform for cloud native network functions (CNF) and virtual network functions (VNF) deployments.
The primary goals are to reduce deployment time and complexity, minimize risks of incompatibility, ensuring application performance is as expected and address customer issues surrounding automation and troubleshooting much earlier in the integration process. The plan is to achieve this by running suites of automated test cases developed and maintained to be in line with industry standards.
This testbed environment will be used to certify real world CNF and VNF configurations, and accelerate their placements. Service providers will benefit from more rigorous and comprehensive accreditation complementary to those associated with Cloud Native and Operator Certifications.
The challenge facing VNFs
VNFs are currently part of a standard network architecture and the service provider journey towards more agile services, but still have limitations. In the initial adoption phase of transitioning physical elements to VNFs, simple physical to virtual machine (VM) migration was done. Without fully decoupling such VMs from underlying infrastructure, this results in single-purpose virtual appliances that are hard to manage and maintain. Additionally, scalability is desired in cloud-native environments, and legacy VNFs make this much harder, if not impossible. This is especially troublesome with NFV being a backbone technology for 5G or edge networks which demand large scale deployment with agility, portability, scalability and lower overhead.
The promise of CNFs
Service provider implementation of virtualization and VNFs began their journey toward cloud-native applications, but realizing full benefits requires an expanded technology approach. Truly cloud-native applications enable loosely coupled systems that are resilient, manageable, observable and with robust automation support.
The adoption of CNFs might prove to be a cure for some of the fundamental limitations of VNFs, by moving many of these functions into containers. Containerization of network architecture components makes it possible to run a variety of services on the same cluster and more easily on-board already decomposed applications, while dynamically directing network traffic to correct pods.
However, CNFs are more than just the containerization of network functions. To get the full benefit of Cloud Native principles beyond container packaging requires further rearchitecting of CNFs, like decomposition, allowing multiple versions during updates, and supporting provided platform services like generic load-balancers or datastores.
Additionally, CNFs and their co-existence and interplay with VNFs - and their legacy systems, are in a necessary transition as adoption of the cloud-native environment extends. As demands escalate, deployment times compress, and complexity of solutions increase, service providers must fully automate the development, deployment, maintenance and operation of the network. Standardized methodologies for configuration and deployment, tools matured in open communities and the ability to perform rigorous testing and certification prior to rollout, are critical for providers now more than ever.
Recognizing the need for a standardized approach, Red Hat and Intel have taken initial methodology and tools developed within OPNFV, and discussed for adoption in Common NFVi Telco Taskforce (CNTT), to create a cloud-based onboarding service in a lab testbed supporting both VNF and CNF environments. It is designed to help mitigate the risks associated with testing and deploying CNFs as well as transitioning existing operational procedures from VNFs to more agile cloud-native methodologies and models across the open hybrid cloud.
Of particular importance is the ability of the testbed to provide a more rigorous certification in the CNF realm. Partners are encouraged to engage with Red Hat to achieve Container and Operator (C&O) Certification for CNFs as required baseline for partners desiring CNF Certification.
CNF Certification subjects participating organizations to a higher degree of rigor and broader scope of testing which is more comprehensive and is specific to demands exhibited by a service provider and expected from a CNF vendor
Virtualization and VNFs began the shift towards NFV, and CNFs continue this transition. More demanding services on the horizon, like 5G and edge computing, will introduce additional complexities as service providers seek to address market and customer needs. This future makes the automation of network deployment and operations critical.
Experience gained in a standardized environment - such as the one being deployed by the cooperative effort of Red Hat and Intel, offer valuable support to smooth the transition from VNF to CNF implementations.
To learn more about the Red Hat Connect Partner program please visit us at https://connect.redhat.com/ and to nominate a network function for certification, please go to https://connect.redhat.com/partner-with-us/nominate-cloud-native-function