This audio was created using Microsoft Azure Speech Services
Previously published on Datacenter Dynamics.
The future of mobile broadband relies on 5G, the latest generation of telco – and this generation is going to rely on a multi-level data center architecture. After years of hype, 5G is becoming a reality with major telco providers finishing test sites and announcing plans to broadly deploy it. AT&T is working to advance its 5G efforts with the “5G Evolution”, which is laying the groundwork for 5G coverage across the country through the expansion of its LTE Advanced network. Verizon, meanwhile, recently announced its first 5G network went live in Chicago in early April.
Still, most carriers remain vague on planned dates to have their 5G networks online, as the majority of operators realize that for 5G to function at its claimed performance level of 10Gbps and <1ms latency, providers must ensure they are laying the foundation for a network of data centers that can support its deployment and operational requirements. This is all to say that in 2019 and beyond, mobile users won’t be aware of the underlying data center architecture, but, they surely feel the effects when uploading or downloading is very slow or the service is not available – when the edge goes down. That’s why we need to rethink resiliency at the edge.
Strategies for Building and Maintaining Resiliency at the Edge
Telecom companies will spend billions to support their demanding mobile edge cloud. Still, edge computing challenges exist, mostly due to the massive number of deployments that need coordination and the corresponding remote and site management needed for upkeep and troubleshooting. As 5G forces the industry toward the edge, we must look for ways to support the growth and maintain the increasingly critical edge by creating a collaborative ecosystem that builds resiliency.
To effectively and efficiently manage edge, operators should consider these three must haves for building and maintaining resiliency:
- An integrated ecosystem: Edge resiliency can only be achieved and maintained when physical infrastructure vendors, system integrators, IT equipment manufacturers and managed service providers work together to create an ecosystem of resources that provides open systems and open access to data and outputs. Giving access to all required stakeholders will be critical for data-sharing. Additionally, this ecosystem of stakeholders needs to work together towards fully integrating modular micro data centers at the edge, which allow providers to quickly build and deploy data center capacity in nearby metros when it’s needed. These environments include all the IT hardware, security, networking, environmental monitoring, rack access and physical infrastructure needed within a stand-alone secure enclosure to protect critical business applications.
- Management tools: Legacy data center tools are no longer adequate for edge deployments. With a lack of staff available at edge locations to manage operations and respond to issues, management tools, especially those that are cloud-based and encourage data access and sharing, keep sites secure and provide advanced warning before issues arise. With a cloud-based management tool, operators will no longer have individualized tools for each site that need to be managed separately and require individual IP addresses. Instead, they’ll have access to an app or website that provides one dashboard to manage all components as a single system at any given edge site, reducing maintenance requirements, enhancing efficiency and providing enhanced cybersecurity to protect against potential threats.
- Analytics and AI: To support actionable decision making, data must be consolidated across the edge network and analyzed to determine potential issues before downtime occurs. By integrating analytics and AI into a solid foundation that’s supported with a cloud architecture, as well as data lake and subject matter experts, data can be properly aggregated, analyzed and secured. Supported with machine learning and AI capabilities, operators can more accurately create AI use cases, identify critical variables, normalize data sets and provide the power needed to analyze all data. By combining the capabilities of AI and analytics with human talent, teams can benchmark performance, determine patterns, generate alerts and create scorecards to minimize hours spent on alarms, reduce downtime and enhance their peace of mind.
Overcoming these Challenges for a Critical Edge
While 5G is not fully deployed, it’s still a major factor driving the market toward the increasingly critical edge. Service providers already feel the pressure to process data quickly and prevent their customers from experiencing the effects of latency. Now, the availability of smaller data centers has a bigger implication. To successfully manage this brave new world, the industry needs to rethink its approach to resiliency at the edge and explore solutions that will help overcome the major challenges to getting there. Check out this full presentation I held at OCPSummit19 where I discuss ensuring resiliency at the edge through a collaborative ecosystem.
