Virtual Network Function Orchestration (VNFO) Market Overview: VMs vs Containers
by Kaustubha Parkhi, edited by Alan J Weissberger
Cloud-native network functions (CNF) promise to change the dynamic of telecommunications network function engineering. The advent of 5G has added impetus to this change. Insight Research is at the cutting edge of CNF market analysis. Here are a few excerpts from our new report on the Virtualized Network Function Orchestrator (VNFO) market
Insight Research considers Virtual Machines (VMs) and Containers to be the major Virtual Network Function Orchestration (VNFO) methodologies. Network functions synthesized using VMs and Containers qualify as Virtual Network Functions (VNFs), in our opinion. That latter term has taken on much broader context since it was first introduced in the context of Network Function Virtualization (NFV) at the OpenFlow World Congress in 2012.
VNFs orchestrated by Containers are sometimes referred to as cloud-native NFs (CNFs). Insight Research has also employed this term as early as 2020. Over time however, we have observed that the usage of CNFs is neither consistent nor uniform.
Most ‘traditional’ Management and Orchestration (MANO) schemes such as ONAP, OSM and all proprietary offerings now support Containers and Kubernetes [1.]. Containers are thus one more mean towards achieving the end-objective of VNFs. As such, Insight Research finds it more appropriate to use VNF as an umbrella term and refer to VM or Container as the specific virtualization methodology.
Note 1. Kubernetes, also known as K8s, is an open-source system for automating deployment, scaling, and management of containerized applications.
The question then arises – where would we slot network functions (NFs) orchestrated by containers encapsulated in VMs? Answer is containers. Similarly, NFs orchestrated by VMs encapsulated in containers are slotted under VMs.
The table below breaks down the VNFO market by virtualization methodology.
We see Containers gaining major market share from VMs such that they are running away with the VNFO market. The advantages of containers over VMs are all well known. Containers are sleeker and when employed with optimal microservice granularity – considerably faster as well.
Additionally, VMs have a head start over containers and have established a solid legacy which will hold good for near to midterm future. However, barriers surrounding container adoption are gradually dissolving with differing momentums across end-applications. The greater the performance differential, the greater is the adoption potential for Containers in end-applications.
Initiatives such Nephio have placed Kubernetes in the center of the VNFO universe. In short, it’s a matter of time before Containers push VMs to be the dominant VNFO virtualization vehicle. However, many question arise.
Here are a few questions for starters:
Is the NFVO the same as service orchestrator?
Is the NFVO the same as SDN controller?
Is Kubernetes an orchestrator?
Since containers and VMs can be embedded inside one another, how do we stamp which virtualization methodology they are using?
If a proprietary MANO uses portions of open source code, should it be considered proprietary?
After picking the brains of numerous experts who were unfailingly patient in unravelling their thinking, Insight Research has been able to arrive at a set of clearcut definitions and assumptions that address the above queries and more.
To buy the report or download an executive summary, please visit:
https://www.insight-corp.com/product/vnfo-ripe-for-change/
3 thoughts on “Virtual Network Function Orchestration (VNFO) Market Overview: VMs vs Containers”
Comments are closed.
It is important to reiterate the reasons for including containers under VNF and in essence, Kubernetes under VNFO. Both VMs and containers are virtualization methodologies. Thus, network functions synthesized using VMs and containers qualify as VNFs.
In VNFs orchestrated by containers are sometimes referred to as cloud-native NFs (CNFs). Insight Research has also employed this term as early as 2020. Over time however, we have observed that the usage of CNFs is neither consistent nor uniform.
Most ‘traditional’ MANOs such as ONAP, OSM and all proprietary offerings now support containers and Kubernetes. Containers are thus one more means towards achieving the end-objective of VNFs.
In such situations, Insight Research finds it more appropriate to use VNF as an umbrella term and under this term, refer to either VM or container as the specific virtualization methodology.
The question then arises – where would we position NFs orchestrated by containers encapsulated in VMs?
Answer is containers.
Similarly, NFs orchestrated by VMs encapsulated in containers are slotted under VMs.
https://www.linkedin.com/pulse/cnfsvnfs-kaustubha-parkhi/?trackingId=
I’ve tried OpenAI’s ChatGPT and was not impressed. The Chatbot said it could not open any webpage I provided and could not remember any conversation in its data base. After logging out then logging in again, I asked about the previous data provided, but the Chatbot did not remember any of it. :-((
Cloud Native Trend #1: Platform teams will drive cloud-native adoption
Just as containers shifted the IT conversation away from virtualized infrastructure, Kubernetes — the decided-on substrate for cloud-native applications — has become the backdrop to the action in this sector as vendors use it to deliver cloud-native services (connectivity, developer tools, security, etc.) across distributed environments. We’re seeing the rise of platform teams as kingmakers in the enterprise, growing traction in the telco realm, a continuing focus on securing the software supply chain and expanded options for application networking (beyond service mesh). While adoption is accelerating, the market remains fragmented, with customers facing a massive amount of vendor and Cloud Native Computing Foundation (CNCF) options. The next wave of adoption will be by companies that don’t have Kubernetes ninja teams, which means better tooling right across the landscape will need to be baked in by necessity. The next steps are to reduce cloud-native complexity and improve integration of its parts to drive increased developer productivity. Additional abstractions, unified APIs, a common dashboard and developer portal/PaaS — all of which are under way to different degrees at commercial vendors and as CNCF projects — will make Kubernetes in cloud easier to use. Meanwhile, the universal ”shift left” into all of the ‘ops’ (e.g., DevOps, SecOps, FinOps, MLOps) demands investment in tools and training to fill the gap between Kubernetes and DevOps teams, a need that can also be met by platform engineers.
Trend #2: Application modernization will remain a difficult part of IT transformation
In less than a decade, cloud nativity has become the prevailing mindset and methodology for application and infrastructure architecture modernization and transformation. Net new and modernized applications are increasingly being deployed on cloud-native platforms. Meanwhile, monolithic legacy applications (and the associated technical debt) endure, limiting enterprises’ ability to reap the full range of clouds agility benefits. Deconstructing and modularizing monolithic code bases is not easy or straightforward. Organizations will increasingly look to automation and managed/ professional services to modernize their application estates to keep the IT transformation train on track.
Trend #3: Cloud-native 5G networks will hit the mainstream phase
The success of 5G at its core will be determined by the extent to which telecom vendors and their customers can master cloud-native technologies and practices. So far, results have underwhelmed during initial rollouts. The next radical shift in network service delivery architecture and approach — to cloud native and DevOps — is necessary to fully take advantage of 5G capabilities such as slicing and network programmability, which will enter the mainstream in 2023. Cloud-native design and deployment principles bring an optimal mix of application portability, reusability, time to market, automation and scale for the next generation of network operations and monetization. 5G core — and by extension edge — are “killer apps” for cloud native.
Source: S&P Global 2023 Trends in Cloud Native