SD-WAN Cloud Edge

VMware SD-WAN: Forming Airtight Connections Across the Omnicloud

The cloud continues to evolve. Many hyperscalers began by providing regional public clouds, which provided cloud service connectivity to regions via VMware SD-WAN.

However, that deployment strategy did not fit all use cases. Hyperscalers responded by adjusting to their customers’ business needs. Harnessing the power of SD-WAN, they brought the cloud closer to users, expanding coverage from whole countries to countless local zones at the near edge.

And when enterprises needed on-premise cloud capabilities at far edge locations to perform edge compute and other roles, hyperscalers provided that too, relying on VMware SD-WAN to connect everything from factory floors to warehouses and more.

Lastly, understanding 5G’s impact, hyperscalers have effectively expanded their cloud services in mobile operators 5g environments thru their public MEC offering.

Similar to how hyperscalers evolved their cloud deployment strategy, businesses are evolving their use of the cloud, aggregating public cloud, MEC, local zones, and edge compute environments (private clouds) — an “omnicloud” approach to achieve their objectives. This empowers them to deploy their services in and across numerous physical footprints, enabling them to provide services that scale across different use cases. And by 2025, according to industry analysts, workloads will be evenly distributed across the private cloud (30%), MEC (40%), and public cloud (30%).

Some companies begin their omnicloud journey by starting small, establishing an edge computing environment as their “home base” cloud. Why adopt that approach? Their primary reason is latency: if they seek speed of light connectivity, nothing beats edge compute performance.

Designed to process workloads instantaneously, most edge compute environments are business-critical virtualized services sitting on premises. But they are not just boxes in buildings, isolated from the world. As enterprise use cases have become more complex, edge compute environments may require connectivity to the omnicloud ecosystem for supporting critical business use cases, ranging from retail to manufacturing.

Let’s explore a few ways that SD-WAN ensures an airtight connection across the omnicloud — effectively connecting edge compute environments with public clouds, local zones, and public MEC.

Inside SD-WAN’s secret sauce for connecting edge compute environments to the omnicloud

Serving as a network overlay, SD-WAN delivers real-time reporting on underlying WAN transports’ performance, immediately making corrections whenever needed so connectivity to the omnicloud never falters. These features act as the bedrock to power reliable connections:

  1. Dynamic Multipath Optimization (DMPO). Network bottlenecks are always troublesome due to the tsunami of apps and data flowing through tunnels that are too small or exhibit erratic behavior. These bottlenecks radically reduce the quality of service, especially for processes that require low latency connectivity and rely on MEC to power super-fast performance. DMPO, a special feature of VMware SD-WAN, examines every possible path to MEC (or the public cloud), reports on each tunnel’s quality of experience, and intelligently reroutes app traffic to the best tunnels — significantly improving user experience.
  • VMware Edge Network Intelligence. The need to improve app experiences and boost user productivity remains at the center of why enterprises shift workloads to different clouds. But as real-time app traffic surges, IT teams struggle to understand the issues that networks constantly face. In response, IT teams use SD-WAN and VMware Edge Network Intelligence to increase connectivity across the omnicloud and rapidly solve problems.

Looking under the hood, ENI delivers a deep dive into network performance indicators, finds root causes, isolates issues, and performs remediation — saving IT teams time while ensuring great end-user experiences.

  • VMware SASE Orchestrator: Managing multiple clouds is complicated. IT engineers cannot be experts in countless interfaces. Performing backend API integration with hyperscalers, the SASE Orchestrator helps engineers perform centralized configuration deployment, monitoring, management, and operation of those interfaces via a single-pane-of-glass.

Additionally, it empowers them with the flexibility to do everything from spinning up a hyperscaler’s service that is connected to a VMware SD-WAN virtual edge device to launching a virtual edge device within a hyperscaler’s portal. And by combining Orchestrator with VMware Edge Network Intelligence, IT teams are armed with incredibly deep visibility into end-user and IoT device experiences.  

Putting the omnicloud to work

How do industry verticals harness the power of omnicloud?

Within the retail world, some enterprise stores offer smart carts, eliminating the need for cashiers. How does this work? Upon exiting the store, its edge compute environment sensor processes your cart’s contents and charges your credit card. But your purchase data does not simply lie dormant. Enterprises leverage it to learn what you might buy in the future.

So, when shoppers return to the store, their smart carts use SD-WAN to securely connect with local zones, which provides them with personalized product suggestions. Intrigued, shoppers may purchase an item, resulting in a win for the company because the local zones provided solid recommendations that resulted in a sale.

In the manufacturing world, many plants rely on private MEC — edge devices deployed on customer premises and tied to private 5G connectivity — to reduce latency and jitter, ensuring a resilient network for the field of machines across their factory floors.

How would this look? Imagine a factory floor full of machines. The plant’s edge compute environment tracks every machine’s performance and if one of the machines falters, the system immediately notifies the plant manager for action.

But what if all the machines’ health could be tracked in real time, where even the slightest performance deviation could be reported to the plant manager for action? In this scenario, the plant would use SD-WAN to connect with Private MEC, which may detect a slight variation in a machine’s operating temperature.

The plant manager may choose to investigate or wait until the temperature reaches an unacceptable degree, thus preventing the snowball from turning into an avalanche.

Meanwhile, plant’s public cloud receives that data, along with data from the enterprise’s other factories within that region, noting deviations within machines that are common across every factory floor — empowering plant managers with actionable information to take corrective measures and potentially replace those machines. In the future, the replacement could be automated, taking that task off managers’ plates.

Learn more


Leave a Reply

Your email address will not be published. Required fields are marked *