Security, Cloud & the SBC

It's hard to deny that 5G has garnered the greatest number of headlines in the first three months of 2017. This was certainly the case at Mobile World Congress (MWC) 2017 a few weeks ago, and the fever pitch of 5G announcements is unlikely to abate anytime soon.

However, my experience at MWC was that while 5G was the "sound bite" leader, behind this backdrop, cloud and security technologies are fundamentally driving business growth opportunities for communications service providers (CSPs). As a result, I was pleasantly surprised at just how quickly these technologies are maturing, as well the high levels of resiliency that existing products have also shown. Let me explain.

Given the whirlwind of cloud R&D that has taken place over the past 12 months, while some challenges remain, in my opinion, the cloud is now recognized as the foundational architecture that will fuel current and future CSP growth prospects. However, while I believe the cloud is now mature enough to support large-scale cross-domain deployments, it is not a static architecture, it is an evolving architecture that will morph in response to new technology requirements. The evidence of this trait was readily apparent at MWC and in other recent discussions I had with vendors and CSPs alike. As a result, I view the dynamic world of the cloud as continuing to redefine a number of key existing network functions.

A number of network functions stand out. The first is the session border controller (SBC). It's probably not a surprise that SBCs are retaining and even gaining relevancy in the cloud domain when you break down the capacities they support and the edge realm in which they operate. SBCs, since inception, were designed to be application-aware, manage both the control-plane and media-plane and enforce security policies at the network edge, so it only makes sense they will have a key role to play in the cloud. However, in order to stay relevant, SBCs are quickly moving to support virtualization configurations and optimized performances of the virtualized media plane to enable large-scale access and peering services rollout.

However, virtualization is not enough, given complex cloud operational requirements. Therefore, as we move to a more distributed cloud edge model, SBCs are being asked to manage a greater number of more complex services – including cloud-native services. This was evident at MWC, with a number of fixed and mobile use cases and demonstrations that focused on leveraging the additional software intelligence of a new generation of cloud vSBCs vs. taking existing products and renewing them to become cloud-optimized.

In terms of specific use cases, on the fixed side, it's clear that virtualizing enterprise services is driving software intelligence to the edge to reduce latency budgets and to also provide end users with greater control on a much larger scale. This means that even established core functions such as IP Multimedia Subsystem (IMS), once virtualized, will also be pushed to the edge to work closely with this latest generation of cloud vSBC. This may not sound like a major change, but from a business perspective, in my view, it is a profound development since it facilitates a harmonized managed service delivery model that doesn’t currently exist.

Essentially, since vSBC and vIMS software-based virtual network functions (VNFs) can be spun up anywhere in the cloud (either centralized core or distributed edge), it becomes possible to orchestrate and seamlessly integrate a suite of services such as WebRTC in-browser communications and unified communications (UC) that were previously constrained by the network boundary and silos in which they executed.

Since the cloud is a software fabric, the notion of boundaries and hence limitations are greatly reduced. In turn, this opens the door for vSBC-equipped networks to manage the orchestration and lifecycle of these services since they support elastic scale on both the control and media plane independently, which is critical to provide the flexibility to harmonize these services and meet low-latency performance targets. This ability to elastically scale and execute services as a single fabric is key to monetization and, in my view, a key reason the managed services model continues to evolve and gain market momentum.

The caveat here, of course, as also quietly discussed in the halls at MWC, is the requirement to adopt a new security model. The greatest challenge I see here is that pre-cloud security implementation was more straightforward in that there were well-defined perimeters where defenses and countermeasures should be placed. In reality, as with the breaking down of services barriers, this is no longer the case, since software VNFs as potential intrusion points are fluid in terms of running in the core or at the edge.

However, thankfully the latest generation of vSBCs now reaching the market supporting pure software implementations are well positioned to play an even stronger role in securing the cloud on both the control and media planes. This was also noted at MWC with a must greater business focus of leveraging the vSBC to support hosted security as a service (SECaaS) add-ons for a wide variety of applications including voice-over-LTE-as-a-service (VoLTEaaS).

The concept that is now emerging to support these services is that vSBC as part of this harmonized service model will work closely with analytics and vIMS to manage application performance, and simplify the ability to introduce on-demand cloud-native value added services (VAS), while also leveraging advanced packet filtering capabilities to mitigate the threat of DDoS attacks.

Looking ahead, I believe the push to support 5G and artificial intelligence (AI) will further enhance the role of vSBCs and their value proposition in executing services and securing the cloud in whatever manifestation(s) the cloud has assumed at that point in time.

References