One of the many promised benefits of 5G is faster and more reliable access to real-world experiences and data, thanks to the elimination of the inevitable lag or latency that can occur when something connects to a wireless network. In fact, fast response times and reduced latency are some of the most popular technical advantages of 5G network architectures and 5G NR (New Radio) signaling specifications.
However, in themselves the pure 5G aspects of latency reduction have been at best at best, especially because some of them depend on the use of 5G SA (stand-alone) networks, which are still being implemented in The United States and around the world (seeImprovements in 5G latency are still lagging behind” for more).
Much of the real magic depends on bringing more computing resources to the brink of wireless networking, a concept that is part of the larger world of finite computing. In fact, as I recently argued (see “Moving 5G forward requires a big pictureFor more), the 5G that we were all promised, along with all the amazing features it had to provide, is essentially impossible without large-scale computing.
Fortunately, after years of little more than talk, we are finally starting to see some critical new technologies on the market that need to start making this possible. At two of the busy technical conferences this week, both Dell Technologies and Google discussed some of their latest advances in computer technology. As each company approaches the market from its own unique perspective, together they use this opportunity to make things like deploying mini-cloud data centers in millions of cell towers and other wireless network presence points look temptingly real.
For its part, Dell Technologies debuted several different hardware, software and service options focused on the edge of its Dell Tech Summit virtual event. Some of these options, such as the company’s validated design with Litmus, are specifically targeted at industry-specific vertical end-applications on site, while others are applicable to wider network end-applications.
From a purely computational point of view, Dell has introduced several different options that can operate in space-constrained and / or challenging environments that network-driven management computing systems often encounter. The new Dell EMC VxRail satellite units reduce the company’s well-established Hyperconverted Infrastructure (HCI) line to a single trunk size, allowing it to fit in a wide variety of locations. As with all VxRails, the new satellite nodes are designed with VMWare and are optimized to work with VMWare-related applications, such as the new VMWare Edge offerings such as the VMWare Edge Compute Stack. As a result, communication service providers, as well as private companies, can run either a virtual machine (VM) or container-based applications in these highly based environments.
For places with more demanding physical requirements, the new Dell EMC Edge Gateway offers a robust, small form factor device that can withstand temperature extremes of
-4 to 140 degrees Fahrenheit, while providing the local calculations and storage needed to perform a wide variety of edge-optimized applications, including analysis, data processing, and more.
On the software side, the Dell EMC Streaming Data Platform can use Nvidia GPUs to perform video-based analysis for end applications on properly equipped VxRail and PowerEdge servers. Bare Metal Orchestrator is designed to allow telecom operators and other service providers to manage the hundreds of thousands of independent, highly distributed end-to-end servers that CSPs will need as they work to upgrade their networks. In particular, the shift to more software-defined and Open RAN (network-based network access networks) creates a demand for huge amounts of general-purpose computing hardware, all of which need to be managed easily and autonomously — tasks that Bare Metal Orchestrator is designed.
At the next Google Cloud event in 2021, the company made a wide variety of messages in many different aspects of its GCP cloud service and workspace offerings, but some had a very clear focus on the final calculations. Google Distributed Cloud Edge and Google Distributed Cloud Hosted leverage the company’s Anthos feature and allow Google Cloud infrastructure and services (as well as the applications that use them) to run remotely.
The interesting thing about Google’s new proposal is that it specifically covers (and calls for) many types of edges and locations (of course, in the process, also emphasizing how potentially confusing the definition of a finite calculation can be …). In particular, Google Distributed Cloud can run across the edge of the Google Network (over 140 sites worldwide), telecommunications companies, and 5G-related “edges” of other service providers, on the edges of customers, or in joint location or other client-owned data centers.
On one level, this is Google Cloud’s effort to expand its private cloud offerings and compete more directly with Amazon AWS’s Outposts and Microsoft’s Azure Stack. From a larger picture, however, the goal seems to be clearly focused on providing an infrastructure to facilitate the deployment of end-specific applications.
Taken together, these announcements from Dell Technologies and Google are beginning to give a sense of momentum to improved capabilities and a focus on end-to-end computing discussed by many different vendors. Clearly, further steps need to be taken, but in the light of this news, the concept of making it easier to create, deploy and manage finite computing resources, applications and services is starting to focus better.
On the one hand, this is great news for telecom operators and other service providers. This gives them better access to the tools they need to modernize their networks, prepares them to offer new services, and even begins to empower them as providers of finite computing – perhaps individual new opportunities and business offerings. . At the same time, increased participation by large technology providers and internal efforts in large enterprises, which the same tools could inspire, could also potentially block carriers from these new opportunities. For example, if IT departments start thinking that they can create something like their own private 5G networks to run manually based applications only with the help of their existing vendors / technical vendor partners, telecom operators may be challenged.
In reality, most organizations will not want to deal with the additional difficulties of building and maintaining private networks. They are likely to partner with telecommunications companies or other solution providers to get the job done. Yet this certainly raises some interesting questions and opportunities. He also suggests that there will be a very large number of countries interested in seeing the world of finite computing adapt and evolve with the advent of new tools.
Disclosure: TECHnalysis Research is a market research and consulting company in the technology industry and, like all companies in the field, works with many technology providers as customers, some of which may be listed in this article.