This post is co-authored by Jason Shepherd, IoT and Edge Computing CTO, Dell Technologies.
Telecommunications service providers are in the midst of a transition that will enable a new generation of applications based on 5G connectivity and edge computing. From the evolution of the central office to backhaul and pre-aggregation facilities, the “Network Edge” represents an opportunity for telecommunications service providers that is ripe for transformation.
Dell Technologies and VMware are excited to welcome our telecommunications service provider customers to VMworld this year and have planned a curated experience that includes:
- Breakout Sessions
- Theater sessions
- TheCube Live
- A Hospitality Suite
- Dedicated 5G booth experience
- And an evening on the San Francisco Giant’s Yacht
This experience will highlight how Dell Technologies and VMware are working together to accelerate our customers’ transformation.
Dell Technologies and VMware have partnered to deliver validated solutions that help telecommunications service providers de-risk NFV infrastructure deployments and accelerate time to ROI. Over time, and through collaboration on open APIs, we will enable advanced capabilities across the mobile network that are context-aware of application quality of service (QoS) requirements. This can only happen with community-wide, open collaboration on abstractions and a commitment to interoperability – from the edge all the way to cloud.
Dell Technologies has been paving the way through our work in EdgeX Foundry and Akraino, and through contributions to the broader Linux Foundation’s LF Edge project. On top of these open foundations, Dell Technologies and VMware aim to deliver the software and hardware that will enable telecommunications service providers to rapidly develop and bring to market new services that support emerging 5G and edge use cases. Our mutual goal here is to help customers architect the edge for success.
Extending the Cloud To The Network Edge
When we think of the cloud, we think first of homogeneous building blocks, replicated as a means of enabling scale. We replicate standard building-block servers within a rack, replicate that rack across a row, then replicate that row across a data center. The challenges of centralized scale – virtualization, security, automation, management, workload mobility – are not only well-understood, but largely resolved.
So why is the network edge so different? For starters, the edge is a heterogeneous environment. The “architect once, scale indefinitely” approach espoused by public clouds doesn’t apply to the edge. The wide variety of locations, facilities and workloads at the edge requires the adoption of a “design to-suit” philosophy. The telecommunications service provider operational model must be flexible enough to accommodate use-case specific infrastructure at the edge. Various degrees of ruggedization and regulatory considerations, as well as any combination of acceleration hardware such as GPUs, FPGAs, Smart NICs, can be incorporated to meet these requirements. With the right architectural principles that include open standards and industry standard hardware, the required heterogeneity of infrastructure can be managed smoothly through software.
The edge does not lend itself towards “standardization,” per se. The ability to replicate homogeneous building blocks at the edge relying on the same tools, technologies, and operating models used in massively-scaled, public cloud data centers is not apparent. The closer and closer you get to the edge, the more unique both hardware and software need to be to meet the demands of specific locations and workloads. Combine this with lights-out, un-manned facilities in a heavily regulated industry, and the need for a “Telco Cloud” has never been more apparent.
The Holy Grail
Management and control of remote facility infrastructure is not as simple as it is in the data center. Edge facilities can be difficult to reach and can have significantly hardened physical security requirements. An early and deliberate investment in management tools that can consistently handle a variety of edge devices will be required. An extension of the NFV model to the edge is also desirable. With NFV, the migration of applications from the core of the network to the edge can be facilitated, enabling new efficiencies. Underutilized resources at the edge can be repurposed quickly for local IT function (e.g. AI/ML at the edge). A loosely coupled, distributed network edge will also provide effective and dynamic service scaling.
Security and privacy concerns must be addressed from the moment data is generated at the edge. In order to scale security, privacy and trust across an inherently heterogeneous system-of-systems that spans public and private boundaries, it’s imperative to start with open interoperability at the edge (A focus of the aforementioned work in the Linux Foundation). This is a core aspect of Dell Technologies’ edge strategy. We’re combining the power of our portfolio with value-add from our global partners, like VMware, on top of an open foundation to help our customers build secure, scalable and trusted solutions.
It is important to start small, but equally important to invest in the right infrastructure to scale big over time. An investment in and commitment to a multi-cloud strategy that starts with an open edge is essential to maintaining control over the services offered and data generated at the edge. After all, where would the Internet that we know-and-love be, without a commitment to open interoperability? Edge computing should follow this model to achieve similar greatness.
Through collaboration on open, trusted infrastructure, we will enable entirely new services and business models. This will ultimately help us realize the Holy Grail of Digital – selling or sharing on your terms data, resources and/or services. Dell Technologies and VMware are prepared to journey with you to 5G and rock the edge!