On Dec. 3rd at AWS re:Invent (Dec. 2-6, 2019) in Las Vegas, Amazon Web Services Inc. (AWS), announced AWS Wavelength, which provides developers the ability to build applications that serve end-users with single-digit millisecond latencies over the 5G network. AWS is partnering with Verizon on making AWS Wavelength available across the United States. Currently, AWS Wavelength is being piloted by select customers in Verizon’s 5G Edge, Verizon’s mobile edge compute (MEC) solution, in Chicago. Additionally, AWS is collaborating with other global telecommunications companies (including Vodafone, SK Telecom, and KDDI) to launch AWS Wavelength across Europe, South Korea, and Japan in 2020, with more global partners coming soon. From Amazon’s AWS Wavelength press release:
AWS Wavelength enables developers to build applications that deliver single-digit millisecond latencies to mobile devices and end-users. AWS developers can deploy their applications to Wavelength Zones, AWS infrastructure deployments that embed AWS compute and storage services within the wireless telecommunications providers’ data centers at the edge of the 5G networks, and seamlessly access the breadth of AWS services in the region. This enables developers to deliver applications that require single-digit millisecond latencies such as game and live video streaming, machine learning inference at the edge, and augmented and virtual reality (AR/VR).
AWS Wavelength brings AWS services to the edge of the 5G network, minimizing the latency to connect to an application from a mobile device. Application traffic can reach application servers running in Wavelength Zones without leaving the mobile provider’s network. This reduces the extra network hops to the Internet that can result in latencies of more than 100 milliseconds, preventing customers from taking full advantage of the bandwidth and latency advancements of 5G.
More from the press release:
Wavelength embeds AWS compute and storage services at the edge of wireless telecommunications providers’ 5G networks, enabling developers to serve use-cases that require ultra-low latency like machine learning inference at the edge, autonomous industrial equipment, smart cars and cities, Internet of Things (IoT), and Augmented and Virtual Reality. Wavelength brings the power of AWS to the edge of the 5G network, so developers can deploy the portions of an application that require ultra-low latency within the 5G network, and then seamlessly connect back to the rest of their application and full range of cloud services running in AWS. AWS customers can now use the same familiar AWS APIs, tools, and functionality they use today, to deliver-low latency applications at the edge of the 5G network, around the world […]
With infrastructure that consists of 69 Availability Zones, in 22 AWS Regions, AWS enables developers to serve end-users with low latencies worldwide. However, emerging interactive applications like game streaming, virtual reality, and real-time rendering require even lower latencies, of single-digit milliseconds to end-users and devices, connected through mobile networks. In addition, use-cases like industrial automation, smart cities, IoT, and autonomous vehicles require data processing to take place close to the source in order to conserve resources like device power and bandwidth. The 5G network is up to 20 times faster than 4G, and can be used to dramatically increase the number of supported devices and shrink network latency for mobile devices. However, even with the arrival of 5G, mobile devices still have to cross multiple network hops when connecting to an application over the Internet. Today, application traffic has to travel from a device to a cell tower to metro aggregation sites to regional aggregation sites and to the Internet before it can access resources running in AWS. These network hops can result in latencies of more than 100 milliseconds. This prevents developers from realizing the full potential of 5G to address low-latency use-cases.
Wavelength addresses these problems by bringing AWS services to the edge of the 5G network, minimizing the latency to connect to an application from a mobile device. With Wavelength, AWS developers can deploy their applications to Wavelength Zones, AWS infrastructure deployments that embed AWS compute and storage services within the network operators’ datacenters at the edge of the 5G network, so application traffic only needs to travel from the device to a cell tower to a Wavelength Zone running in a metro aggregation site. This removes a lot of the latency that would result from multiple hops between regional aggregation sites and across the Internet, which enables customers to take full advantage of 5G networks. Wavelength also delivers a consistent developer experience across multiple 5G networks around the world, and allows developers to build the next generation of ultra-low latency applications using the familiar AWS services, APIs, and tools they already use today – eliminating the need for developers to negotiate for space and equipment with multiple telecommunications providers, and stitch together application deployment and operations through different management interfaces, before they can begin to deploy their applications.
AWS Wavelength combines the power of the AWS cloud with the cutting-edge 5G networks of leading telecommunications providers like Verizon, Vodafone, KDDI, and SK Telecom to unlock a new wave of innovative applications and services around the world. By delivering these new capabilities, Wavelength enables developers to serve mobile users with single-digit millisecond latency and to optimize their applications by processing data closer to its source, enabling use-cases across a wide range of platforms – from factories to stores to cars to homes.
To deploy their application to the 5G edge, developers can simply extend their Amazon Virtual Private Cloud (VPC) to include a Wavelength Zone and then create AWS resources like Amazon Elastic Compute Cloud (EC2) instances, Amazon Elastic Block Storage (EBS) volumes, and AWS Elastic Container Service (ECS) and Amazon Elastic Kubernetes Services (EKS) containers. In addition, developers can continue to use familiar and powerful AWS services to manage, secure, and scale their applications like AWS CloudFormation, AWS Identity and Access Management (IAM), and AWS Auto Scaling. This enables developers to easily run a wide variety of latency-sensitive workloads like analytics, IoT, machine learning, game streaming, and AR/VR.”
With Wavelength, we bring 5G and cloud together to give our customers the powerful new capability to run cloud services consistently within a few milliseconds of mobile end-users,” said Matt Garman, Vice President Compute Services, AWS. “This is a game changer for developers that is going to unlock a whole new generation of applications and services. We are really excited to see our customers innovate with these unique new capabilities that they did not have access to before.
AWS Wavelength: Partner testimonials:
From the AWS press announcement, here are current Amazon’s Wavelength partners’ testimonial statements:
“Verizon is building the most powerful 5G network in the U.S. Launched in April, Verizon’s 5G Ultra Wideband network is currently live in 18 cities as well as 16 sporting and entertainment arenas across the country, and plans to expand to more than 30 U.S. cities by the end of this year. “Continuing our tradition of bringing new technology to market first we are excited to launch a mobile edge compute service — integrating our 5G Edge platform with Wavelength to allow developers to build new categories of applications and network cloud experiences,” said Kyle Malady, CTO of Verizon. “Bringing together the full capabilities of Verizon’s 5G Ultra Wideband network and AWS, we unlock the full potential of our 5G services for customers to create applications and solutions with the fastest speeds and ultra-low latency.”
Verizon and AWS will integrate AWS – Wavelength with 5G Edge so developers can begin testing applications on ultra-low latency networks. The plan is to connect 5G applications to AWS cloud services without the hops. The two companies will bring compute and storage closer to 5G users. Deployments are planned in Chicago for select customers in 2020 with additional locations added throughout the year.
“Varjo Technologies Oy is based in Helsinki and is creating the world’s best hardware and software for groundbreaking VR/AR/XR computing devices, merging the real and digital worlds seamlessly together in human-eye resolution. “Simulating things at the same acuity you see in real life is a game changer compared to standard VR approaches. Varjo’s unique human-eye resolution technology helps professionals save time, money, and effort,” said Niko Eiden, Founder and CEO, Varjo. “Not too far down the road, our technology will be fully wireless, collaborative mixed reality. And this workspace of the future needs to be rendered in the cloud – with millions of pixels of extremely high-resolution, uncompressed content with single-digit millisecond latencies delivered to our devices – whether on premises at carmakers or in remote sites, through 5G. Now, instead of having to develop expensive local computing services that would be impossible to run on a battery-operated device, we can use edge computing to scale the rendering power and the business of our industrial-grade VR/MR from thousands to hundreds of thousands of units. Having access to the power of the AWS Cloud, together with 5G’s high bandwidth, low latency, and increased connectivity, is vital to our ability to deliver professional immersive computing experiences and to grow our business.”
“Mapbox is the location data platform for mobile and web developers, providing building blocks to add location features like maps, search, and navigation into any experience and changing the way people move around cities and explore the world. Mapbox tools are used by more than 1.7 million live location developers to power daily experiences for people, technology, and business. “Everyone needs maps, so 600 million people touch Mapbox every month as they read the headlines of the New York Times, check the weather on Weather.com, and find great restaurants or concerts on Facebook,” said Eric Gundersen, CEO and Co-Founder, Mapbox. “Our map gets smarter every time someone touches it, using AI to constantly update traffic and new streets — AWS Wavelength’s ultra-low-latency compute can help us process billions of sensor data updates into better maps by identifying new roads as they’re built, routing drivers around traffic jams, and spotting road construction with the Vision SDK. AWS Wavelength can reduce our refresh timelines from minutes to seconds, delivering Mapbox users a truly living map.”
“SK Telecom, the largest mobile operator in Korea, with nearly 50 percent market share, has been leading the global mobile industry through constant innovations in technologies and services. As a 5G pioneer, SK Telecom is also one of the first telco providers to launch commercial 5G mobile-edge computing (MEC) in collaboration with AWS. “By combining the strengths of SK Telecom’s 5G network and AWS cloud, we are set to bring innovative changes to all individuals, businesses and industries. This collaboration enables exciting use cases like game streaming, headless robotics, Ultra High Definition interactive media, autonomous driving, and smart factories. For example, through the application of AWS Wavelength and SK Telecom’s advanced 5G solutions, a smart factory can enhance the response time of robots performing maintenance, security, and manufacturing tasks, allowing the factory to scale operations without increasing costs,” said Ryu Young-sang, Vice President and Head of MNO Business, SK Telecom. “SK Telecom and AWS are deploying 5G multi-access cloud services at the edge, helping third-party developers and enterprises improve quality of experience, create business models, and accelerate time to market for new revenue opportunities. With SK Telecom’s 5G network, we can jointly develop sophisticated cloud services that can create greater value for enterprises of any size in Korea.”
“Vodafone Group is one of the world’s leading telecoms and technology service providers, with extensive experience in connectivity, convergence and the Internet of Things, as well as championing mobile financial services and digital transformation in emerging markets. Vodafone Business and AWS will provide multi-access edge computing capabilities to developers, Internet of Things (IoT), devices and end users by bringing the AWS cloud closer to the devices that need it, and running AWS Wavelength in strategic locations within Vodafone’s 5G network. “With Europe’s largest 5G network across 58 cities and as a global leader in the Internet of Things (IoT) with over 90 million connections, Vodafone is pleased to be the first telco to introduce AWS Wavelength in Europe,” said Vinod Kumar, CEO of Vodafone Business. “Faster speeds and lower latencies have the potential to revolutionize how our customers do business, and they can rely on Vodafone’s existing capabilities and security layers within our own network.”
“KDDI, a leading telecommunications provider in Japan, offers services that include both mobile and fixed-line communications, and Internet services. KDDI, which plans to launch commercial 5G services in Japan by March 2020, is actively developing its 5G network to enable enhanced Mobile Broadband in both densely populated metropolitan areas and rural areas. “In preparation for our 5G service launch, KDDI has been successfully proving that 5G can be delivered with reliable service quality in Japan in metropolitan and rural locations. We have achieved successful trials, like 5G handovers for high-speed racing cars and trains, a real-time, free-viewpoint video stream at a baseball stadium, and 4K video communication at a major station,” said Makoto Takahashi, President, KDDI. “Having the power of the AWS cloud processing and storage services available at the edge of the KDDI 5G network enables us to accelerate IoT innovation for applications like high-definition VR video streaming, VPS (visual positioning service), smart factories, autonomous vehicles, and more. AWS Wavelength provides Japanese businesses and consumers immediate access to these services over the KDDI 5G network. This will also enable us to address some of Japan’s pressing societal issues, such as revitalizing economies in areas facing population decline, rebuilding infrastructure, and improving prompt reaction to natural disasters.”
The tight coupling of 5G networks with edge computing raises challenging interoperability questions. Currently, it appears that a U.S. end user wanting to use an application that relies on a AWS Wavelength Zone would have to be a Verizon 5G Edge customer with a Verizon 5G end point device. That’s not how we usually think about “the cloud,” which today can be accessed from a wide range of different vendor devices over a wide range of connectivity providers.
At MWC 2019, A&T announced it was working with Microsoft Azure to bring network edge computing (NEC) closer to the end point. We wrote in February 2019 that AT&T is using drones to test the network edge compute capabilities with Azure, working with Israel-based startup Vorpal in its foundry in Plano, TX. Microsoft provided new details of its Azure – AT&T 5G partnership on November 26th:
Microsoft Azure cloud services are being integrated into AT&T network edge locations (closer to customers). This means AT&T’s software-defined and virtualized 5G core – what the company calls the Network Cloud – is now capable of delivering Azure services. NEC will initially be available for a limited set of select customers in Dallas. Next year, Los Angeles and Atlanta are targeted for select customer availability.
That implies if you are an AT&T 5G customer you will, at some point in time, likely have access to Microsoft Azure cloud services via NEC. However, AT&T customers won’t be able to access NEC for any other cloud provider, i.e. AWS, Google Cloud, etc.
Hence, you only get the advantages of edge computing (with much lower latency) if you are locked in to a pair of 5G network and cloud providers that have an edge computing partnership.
And what about roaming or truly mobile, such that a 5G endpoint device (in a train, car, bus, ship, etc) moves from one Wavelength Zone or 5G network to another? Will there be any sort of hand-off between providers and will the 5G device be able to operate on more than the 5G network it subscribed to?
Once again, this issue can only be solved once the complete suite of IMT 2020 standards are finalized and implemented by 5G network operators and endpoint device makers!
Verizon 5G Overview:
Verizon’s 5G network strategy is centered on three deliverables with fiber optics for backhaul playing a huge role in all of them:
- 5G mobile for businesses and consumers,
- 5G home broadband (see Note 1. below) —delivering home internet over the air—and
- Mobile edge computing, which is essentially miniature data centers distributed throughout the network so they’re closer to the 5G endpoints.
The company’s CEO Hans Vestberg said that a total of 30 5G mobile cities will be launched by Verizon this year. He also plans to restart Verizon’s fixed wireless 5G Home service  later this year. 5G Home currently is in four U.S. markets.
Note 1. There is no standard for 5G fixed wireless and none is even being worked on. It is not an IMT 2020 use case within ITU.
Fiber and Mobile Edge Computing:
The U.S.’s #1 wireless carrier by subscribers will continue to install fiber at a rate of 1,400 miles per month in support of its 5G network builds for between two and three years. Verizon will begin to provide mobile edge computing [aka Multi-access edge computing (MEC)] during the upcoming quarter, Vestberg said at a Goldman Sachs Communacopia investor conference on Thursday, September 19th. Verizon fiber deployments are critical to supporting a mixture of services, Vestberg said.
As part of its Fiber One project, two years ago Verizon signed a $1.1 billion, three-year fiber and hardware purchase agreement with Corning to build a next-generation fiber platform to support 4G LTE, 5G, and gigabit backhaul for 5G networks and fiber-to-the premise deployments to residential and business customers. Also in 2017, Verizon also announced a $300 million fiber deal with Prsymian Group to provide additional fiber for its wireline and wireless services.
“The whole Intelligent Edge Network was basically all of the way from the data center to the access point we have one unique network for redundancy. And then, of course, in between fiber to the access point and then you decide if its 5G, 4G, or fiber to the home or fiber to curb, or fiber to the enterprise,” Vestberg said. “In that, the fiber deployment for us was extremely important.”
“One part of the whole intelligent edge network was that . . . all the way from the data center to the access point you have one unique network with a lot of redundancy and, in between, a lot of fiber to the access point and then you decide if it’s 3G, 5G, 4G or fiber to the home or fiber to the curb or fiber to the enterprise,” he explained.
Vestberg said: “You have one unique network with a lot of redundancy and, in between, a lot of fiber to the access point,” he said of edge computing, which has become a priority for many wireless and wireline network operators.
mmWave for 5G:
Verizon will continue to deploy millimeter wave (mmWave) for its 5G network for the foreseeable future, Vestberg told the investor conference audience. High frequency band mmWave has great download speeds but its range is very limited, which requires many more small cells.
“Maybe you have 50 to 70 megabits per second on a 4G network today, when you get 1 gig [on 5G] it’s a totally different experience and what you can do with it,” Vestberg said. “What we saw in the 4G era was enormous innovation coming with that [greater] coverage and that speed [over 3G]. It’s going to be the same with 5G for sure,” he added.
“Now we have 2 gigs [gigabits per second] on the phones,” Vestberg said. The range, however, can veer from 2,000 feet to 500 feet and the network can’t deliver flashy streaming videos — or, in fact, any kind of service — indoors. Verizon is the only US carrier solely dedicated to the highband (28GHz) approach to 5G for now. AT&T and T-Mobile plan to launch low and mid band 5G networks next year, along with limited mmWave deployments. Sprint has mid band 5G launched so far.
“We can launch nationwide with millimeter wave,” the Verizon CEO insisted. “Any spectrum will have 5G in the future,” Vestberg noted. Verizon will also offer dynamic spectrum sharing (DSS) in the future. DSS will allow operators to share spectrum instantaneously and simultaneously between 4G and 5G networks. But not for mmWave, since that doesn’t share spectrum with any 4G networks.
Vestberg said Verizon has all the spectrum it needs now to do a nationwide network on mmWave, and that adding more antennas in a given area or making software adjustments are also options for increasing capacity on existing spectrum bands.
Vestberg insisted that the mmWave-based service will be “self-install.” This would be more economical than the “white glove” — a.k.a. professional — installation model that 5G Home started with in October 2018.
Verizon’s mobile network:
A growing percentage of Verizon’s mobile subscribers are on unlimited data plans, with about half today. “This is a way for us to continue to see that our customers have a great journey from metered plan to Unlimited (data) plans and then they can move up…to 5G,” Vestberg said.
“We think that we are best equipped to leverage the best network and continue to partner with [media companies] rather than us managing it. Others might have better qualities for doing that but we don’t, Vestberg said.
Verizon to speak at Goldman Sachs Communacopia Conference September 19
Andrew Dugan, senior vice president and chief technology officer, CenturyLink, Inc. presented his company’s views on network virtualization and related topics at the Cowen and Company 5th Annual Communications Infrastructure Summit in Boulder CO., on Aug. 13th. You can listen to the audio webcast replay here.
Dugan said he doesn’t know what AT&T means when the mega carrier says it’s virtualizing 75% of its core network by the end of 2020. “I’d like to figure out what AT&T means by 75% virtualization,” said Dugan. “I don’t get it. The concept of virtualizing the core router or an optical platform, that’s a lot of cost of your network to provide services. We’re not working on virtualizing that stuff.”
Dugan said CenturyLink is focused on virtualizing systems that enable its customers to turn up and turn down services on demand, and it’s also focused on virtualization at the edge of its network. He said the company likes the benefits of putting a white box device on the customer premises and “letting a customer turn up a firewall or an SD-WAN appliance or a WAN accelerator whenever they want.”
Earlier this week, CenturyLink announced the rollout of its edge compute-focused strategy, beginning with a several hundred-million-dollar investment to build out and support edge compute services. This effort – which includes creating more than 100 initial edge compute locations across the U.S., and providing a range of hybrid cloud solutions and managed services – enables customers to advance their next-gen digital initiatives with technology that integrates high performance, low-latency networking with leading cloud service provider platforms in customized configurations.
“Customers are increasingly coming to us for help with applications where latency, bandwidth and geography are critical considerations,” said Paul Savill, senior vice president, product management, CenturyLink. “This investment creates the platform for CenturyLink to enable enterprises, hyperscalers, wireless carriers, and system integrators with the technology elements to drive years of innovation where workloads get placed closer to customers’ digital interactions.”
This expansion allows businesses and government agencies to leverage a highly diverse, global fiber network with edge facilities designed to serve their local locations within 5 milliseconds of latency. With this infrastructure, companies will be able to complete the linkage from office location to market edge compute aggregation to public cloud and data centers with redundant and dynamically consumable network.
“Digital transformation is gaining momentum as enterprises across all verticals look to technology to improve operational efficiency and enhance the customer experience,” said Melanie Posey, Research Vice President and General Manager at 451 Research. “As business processes become increasingly distributed, data-intensive, and transaction-based, the IT systems they depend on must be equally distributed to provide the necessary compute, storage and network resources to far-flung business value chains.”
Dugan said the edge compute platform plays into the company’s virtualization efforts, allowing customers the ability to turn up and turn down Ethernet services, increase capacity, change vLANs, and configure their services on-demand.
“That, to me, is where NFV and SDN comes in. We haven’t put a number on the percent of the network. We’re more focused on that customer enablement,” he said.
“When you build out an NFV platform, you’ve got the cost of the white box, you have the cost of the management or virtualization software that runs within the white box, and you have the cost of the virtual functions themselves. If you’re running one or two applications on premise, it’s not cheaper. The real value from NFV comes in the flexibility that it provides you to be able to put a box out there and be able to turn up and turn down services. It’s not a capex reduction…It’s a reduction in operating costs because you’re not having to roll trucks and put boxes out,” Dugan added.
CenturyLink says its “thousands of secure technical facilities combined with its network of 450,000-global route miles of fiber, expertise in high-performance cloud networking, and extensive cloud management expertise make this investment in the rapidly emerging edge compute market a natural evolution for the company.”
Key Facts (source: CenturyLink):
- CenturyLink today connects to over 2,200 public and private data centers and over 150,000 on-net, fiber-fed enterprise buildings.
- CenturyLink’s robust fiber network is one of the most deeply peered and well-connected in the world, with over 450,000 route-miles of coverage.
- CenturyLink is expanding access to its services by expanding network colocation services in many key markets to enable customers and partners to run distributed IT workloads close to the edge of the network.
IHS Markit Survey: Cloud and mobility driving new requirements for enterprise edge connectivity in North America
By Matthias Machowinski, senior research director, IHS Markit, and Joshua Bancroft, senior analyst, IHS Markit
- By 2019, 51 percent of network professionals surveyed by IHS Markit will use hybrid cloud and 37 percent will adopt multi-cloud for application delivery.
- Bandwidth consumption continues to rise. Companies are expecting to increase provisioned wide-area network (WAN) bandwidth by more than 30 percent annually across all site types.
- Data backup and storage is the leading reason for traffic growth, followed by cloud services
- Software-defined WAN (SD-WAN) is maturing: 66 percent of surveyed companies anticipate deploying it by the end of 2020.
- Companies deploying SD-WAN use over 50 percent more bandwidth, than those who have not deployed it. Their bandwidth needs are also growing at twice the rate of companies using traditional WANs.
Based on a survey of 292 network professionals at North American enterprises, IHS Markit explored the evolving requirements for enterprise edge connectivity, including WAN and SD-WAN. The study revealed that enterprise IT architectures and consumption models are currently undergoing a major transformation, from servers and applications placed at individual enterprise sites, to a hybrid-cloud model where centralized infrastructure-as-a-service (IaaS) complements highly utilized servers in enterprise-operated data centers. This process allows organizations to bring the benefits of cloud architectures to their own data centers – including simplified management, agility and scalability – and leverage the on-demand aspect of cloud services during peak periods. Respondents also reinforced the viewpoint that the hybrid cloud is a stepping stone to the emerging multi-cloud.
Changing business demographics is sparking the trend of more centralized applications: enterprises are moving closer to their customers, partners, and suppliers. They are adding more physical locations, making mobility a key part of their processes and taking on remote employees to leverage talent and expertise.
Following the current wave of application centralization, certain functions requiring low latency will migrate back to the enterprise edge, residing on universal customer premises equipment (uCPE) and other shared compute platforms. This development is still in its infancy, but it is already on the radar of some companies.
Hybrid cloud is an ideal architecture for distributed enterprises, but it is also contributing to traffic growth at the enterprise edge. Extra attention must be paid to edge connectivity, to ensure users don’t suffer from slow or intermittent access to applications. Performance is a top concern, and enterprises are not only adding more WAN capacity and redundancy, but also adopting SD-WAN.
The primary motivation for deploying SD-WAN is to improve application performance and simplify WAN management. The first wave of SD-WAN deployments focused on cost reduction, and this is still clearly the case, with survey respondents indicating their annual mega-bits-per-second cost is approximately 30 percent lower, with costs declining at a faster rate than in traditional WAN deployments. These results show that SD-WAN can be a crucial way to balance runaway traffic growth with budget constraints.
SD-WAN solutions not only solve the transportation and WAN cost reduction issue, but also help enterprises create a fabric for the multi-cloud. Features like analytics to understand end-user behaviour, enhanced branch security and having a centralized management portal all make SD-WAN an enticing proposition for enterprises looking to adopt a multi-cloud approach.
This IHS Markit study takes explores how companies are advancing connectivity at the enterprise edge, in light of new requirements. It includes traditional WAN and SD-WAN growth expectations, growth drivers, plans for new types of connectivity and technologies, equipment used, feature requirements, preferred suppliers, , and spending plans.
Edge computing will get its primary propulsion from demand for video services, IHS Markit found in a survey. The Linux Foundation commissioned IHS Markit to identify the top apps and revenue opportunities for edge compute services. Video content delivery was cited by 92% of respondents as the top driver of edge computing, while augmented/virtual reality, autonomous vehicles and the industrial internet of things (IIoT) all tied for second place.
During a keynote address at this week’s Layer123 SDN NFV World Congress at The Hague, IHS Markit’s Michael Howard, executive director research and analysis, carrier networks (and a long time colleague of this author), presented some of the results from the market research firm’s survey of edge compute application survey respondents.
“The edge ‘is in’ these days in conversations, conferences and considerations—and there are many definitions,” Howard wrote in an email to FierceTelecom. “Our conclusion is that there are many edges, but as an industry, I believe we can coalesce around a time-related distance to the end user, device or machine, which indicates a short latency, on which many edge applications rely. The other major driver for edge compute is big bandwidth, principally video, where caching and content delivery networks save enormous amounts of video traffic on access, metro, and core networks.”
IHS Markit defined edge compute as being within 20 milliseconds of the end user, device or machine. When compared to Internet Exchanges, telcos have an advantage at the edge because they are much closer to the users via their central offices, cell sites, cell backhaul aggregation, fixed backhaul and street cabinets.
Integrated communications providers and over-the-top providers have partial coverage for edge compute with distributed data centers that are within the 20 milliseconds to 50 milliseconds range, while telcos can hit 5 milliseconds to 20 milliseconds.
Among the top services that are driving edge compute, video content delivery, which included 360 video and venues, was first at 92% followed by a three-way tie among autonomous vehicles, augmented reality/virtual reality and industrial internet of things/automated factory all at 83%. Gaming was next at 75%, with distributed virtualized mobile core and fixed access in another tie with private LTE at 58%.
Other findings from the survey:
- Surveillance and supply chain management each garnered 33%, while smart cities was last at 25%.
- When it comes to which edge services will garner the most revenue, distributed virtualized mobile core and fixed access, private LTE, gaming, video content delivery and industrial IoT all tied at the top of the survey results.
- Supply chain management, autonomous vehicles and AR/VR tied in the next grouping while surveillance and smart cities tied for last.
- Consumer-driven revenue at the edge includes gaming and video content delivery networks while enterprise-driven revenues will include private LTE, industrial IoT and supply chain.
- Overall, many of the edge deployments will initially be justified by cost savings first followed by revenue-bearing applications.
- Edge compute apps will start out in limited or contained rollouts with full deployment taking years and investments across several areas, according to the survey.
Although edge compute brings services closer to end users and alleviates bandwidth constraints, it’s complex. Even a single edge compute location is complex with elements of network functions virtualization, mobile edge computing and fixed mobile convergence technologies that can spread across hundreds of thousands locations.
There are also authorization, billing and reconciliation issues that need to be addressed across various domains, which could be resolved using blockchain to create virtual ledgers.
Further, there’s a long investment road ahead to fully deploy edge compute. Areas that comprise the top tier of investments for edge compute include multi-access edge compute, integration, edge connectivity (two-way data flows, SD-WAN services, low latency and bandwidth), 5G spectrum and engineering.
Earlier this year, AT&T Foundry launched an edge computing test zone in Palo Alto, California, to kick the tires on AR, VR and cloud-driven gaming. As part of the second phase, AT&T Foundry is expanding its edge test zone footprint to cover all of the San Francisco Bay Area, allowing for increased application mobility and broader collaboration potential.
Sales of network service provider Edge Routers and Carrier Ethernet switches declined 8% year over year in the first half of 2018 (1H-2018), according to Dell’Oro Group’s just released market research report.
“The confluence of tepid telecom spending, maturing 4G mobile backhaul deployments, and new product introductions have contributed to the reduction in demand for routers and switches,” said Shin Umeda, vice president at Dell’Oro Group. “Telecom operators in the US have pared down their spending well below historical levels as they evaluate new technologies and architectures for 5G backhaul, and on top of that, the massive network buildouts in China are slowing after years of growth.”
“On the positive side, Cisco, Juniper, and Nokia have introduced major upgrades to their edge router portfolios that should bring customers back in the coming quarters,” added Umeda.
Cisco announced enhancements to its upgrades ASR 9000 edge routing platform earlier this month, while Juniper Networks has upgraded its MX Series for the demands of 5G wireless networks and announced 400 Gigabit Ethernet which must be primarily for high performance cloud data centers and possibly data center interconnects.
Cisco, Huawei, Nokia (via Alcatel-Lucent), and Juniper Networks were the carrier switch/router vendor market share leaders in the second quarter, according to Dell’Oro. A bounce in European switch/router sales was not enough to fully offset declines in North America and Asia Pacific.
The Dell’Oro Group “Router & Carrier Ethernet Switch Quarterly Report” covers the service provider core and edge router, Carrier Ethernet switch, and enterprise router markets for current and historical time periods. The report includes qualitative analysis and statistics for vendor market shares, revenue, average selling prices, and unit and port shipments.
About the Report:
The Dell’Oro Group Router & Carrier Ethernet Switch Quarterly Report offers complete, in-depth coverage of the Service Provider Core and Edge Router, Carrier Ethernet Switch, and Enterprise Router markets for current and historical time periods. The report includes qualitative analysis and detailed statistics for vendor market shares, revenue, average selling prices, and unit and port shipments. To purchase these reports, please contact us by email at [email protected].
A new study titled, ‘Global Carrier Ethernet Switch and Router Market’ added by Reports Monitor is a collection of different segmentations, which includes the product types and uses, foundation elements of (Market/Product Name), and expected adoption timeline of Carrier Ethernet Switch and Router sales in the industry. This report provides a detailed analysis of the various use cases and applications of Carrier Ethernet Switch and Router Market in the global industry, their benefits, and challenges for their implementation. Moreover, it provides the potential revenue generation from the adoption of Carrier Ethernet Switch and Router for each application over the forecast period.
The market has been studied on the basis of various macro- and microeconomic factors influencing it. While focusing on the major driving and restraining factors for the market, the report also provides an in-depth study of the competitive landscape, investment scenario, regulatory framework, and the key strategies and developments taking place within the Global Carrier Ethernet Switch and Router Market.The report offers a comprehensive evaluation of the market. It does so via in-depth qualitative insights, historical data, and verifiable projections about market size. The projections featured in the report have been derived using proven research methodologies and assumptions. By doing so, the research report serves as a repository of analysis and information for every facet of the market, including but not limited to: Regional markets, technology, types, and applications. Our business offerings represent the latest and the most reliable information indispensable for businesses to sustain a competitive edge.
Enquire for the Discount on this Report at:https://www.marketinsightsreports.com/reports/0914839668/global-and-southeast-asia-carrier-ethernet-switch-and-router-industry-2018-market-research-report/discount?source=thefairreporter&mode=03
IDC Directions is the market research firm’s annual conference, which always delivers an informative and actionable overview of the issues shaping the information technology, telecommunications, and consumer technology markets. IDC speakers look at the current state of various markets, cutting edge trends and future IT developments that are likely to result in transformation and change.
This year’s event only had one session on networking which we cover in detail in this article. A total event summary is beyond the scope of the IEEE ComSoc techblog.
As the edge plays host to a growing array of new applications, the focus ultimately turns to edge networking, which must deliver the requisite connectivity, bandwidth, low-latency, and network services for both enterprise and service provider deployments. Indeed, as IoT and other edge services proliferate, a one-size-fits-all approach to edge networking and network security will not suffice. In this session, Brad Casemore of IDC examined the diversity of network requirements and solutions at the edge, covering physical, virtual, and network-as-a-service (NaaS) use cases and application scenarios.
According to IDC, the “Intelligent Edge” includes both the IT Edge (IT activities performed outside the data center, but within purview of IT) and the OT/Operations Technology Edge (embedded technologies that do not directly generate data for enterprise use, and are outside the direct purview of IT).
That’s in contrast to the “Core,” which is the “IT Data Center” — an information aggregation facility that is located on the firm’s own physical premises, off-premises in a collocation facility, or off-premises at a virtual location such as a public cloud.
Networking at the Intelligent Edge involves three types of sub-networks:
▪ Enterprise Cloud IT Edge (branch networking for the cloud)
▪ Enterprise Branch IT Edge (the evolution of networking at branch offices/remote sites)
▪ IoT Edge (networking to, from, and at the IoT/OT Edge)
Networking provides essential connectivity and bandwidth, but it also provides valuable network and security services that accelerate and optimize application and service performance at the edge. Brad said that significant innovation is occurring in edge networking which are enabling better business outcomes at the intelligent edge. Some examples of innovation are:
• Software Defined Networks (SDN)/Intent-based
• Overlay networks (such as SD-WANs)
• Network Virtualization (NV)/Network Function Virtualization (NFV)
• Network security (software-defined perimeter)
As a result, the intelligent edge network is significantly contributing to automated network intelligence, in addition to providing wireless and wireline connectivity services.
Enterprise IT is being challenged to provide access to public and private clouds while also maintaining secure and effective communications with regional offices and headquarters (usually through an IP-MPLS VPN).
Enterprise WANs are not effective for Cloud access, because they lack agility, flexibility, and efficiency.
These two issues are depicted in the following two IDC figures:
As a result, a different Application Centric WAN architecture is needed. Brad proposed SD-WANs for this purpose, despite the reality there is no standard definition or functionality for SD-WAN and no standards for multi-vendor inter-operability or inter-SD-WAN connections (e.g. UNI or NNI, respectively). SD-WAN is an overlay network that provides user control via the Application layer, rather than via a “Northbound” API to/from the Control plane (as in conventional SDN).
The use cases for SD-WAN have been well established, including improving application performance by enabling use of multiple WAN links, simplifying WAN architecture, reducing reliance on MPLS, and improving SaaS performance by automatically steering traffic based on application policy instead of back hauling all traffic to the data center.
IDC believes the Internet of Things (IoT) will have a huge impact on networking infrastructure, especially at the edge where low latency/ real time control of IoT devices will be needed.
Casemore said that SD-WANs will help companies overcome issues associated with a traditional enterprise WAN, which wasn’t built for cloud and lacks operational efficiently.
In a real world example of SD-WANs for a medical device supplier, Brad noted the goals were:
• Dynamic access to all available bandwidth (underlays)
• Move away from using relatively expensive MPLS circuits for voice traffic
• Prioritize business-critical cloud apps ahead of nonproduction apps/traffic
• Need for greater visibility –quickly remediate issues and respond to evolving application/service needs
Benefits cited were the following:
• Improved resilience
• Better application performance and availability
• Cost-effective bandwidth utilization
• Better visibility (faster troubleshooting/remediation and proactive planning)
• IT department and network team now contributing to the business of making and shipping products quickly
IDC sees SD-WAN evolving to incorporate more intent based networking and intelligent automation, with business intent consistently applied to application delivery and performance, he said.
Editor’s Note: Intent-based networking is a hot buzzword in the industry right now, generally describing technology that uses automation and machine learning to implement business policy with little or no human intervention. Many believe that intelligent automation will be how business intent is applied to application delivery and network performance across the WAN.
Brad also suggested the following additional attributes for future SD-WANs
▪ Machine Learning and AI – SD-WAN must become cognitive, proactive, and ultimately self-driving, continuously adapting to changing conditions
▪ Pervasive Security – Applications automatically steered over appropriate links and to appropriate security devices. Secure segmentation provided on a per -application basis.
▪ Stepping stone toward SD-branch
“This is all moving us toward the software-defined (SD)-branch. SD-WAN serves as the precursor and serves as the essential conduit to SD-branch and network as a service (NaaS) at the edge,” Casemore said.
In the SD-branch, routing, firewall, and WAN optimization are provided as virtual functions in a cloud-like NaaS model, replacing expensive hardware. Management is automated and services can be easily adjusted as business needs change, Casemore said.
IDC believes telcos will use SD-branch to provide virtual CPE and unversal CPE services as per this slide from IDC:
Sidebar on SD-branch from a recent Network World article:
The branch network is a critical piece of the IT infrastructure for most distributed organizations. The branch network is responsible for providing reliable, high quality communications to and from remote locations. It must be secure, easy to deploy, able to be managed centrally and cost effective. Requirements for branch networks continue to evolve with needs for increased bandwidth, quality of service, security and support for IoT.
SDN and network virtualization technologies have matured to the point where they can deliver significant benefits for branch networks. For example, SD-WAN technology is rapidly being deployed to improve the quality of application delivery and reducing operational complexity. SD-WAN suppliers are rapidly consolidating branch network functions and have reduced (or eliminated) the need for branch routers and WAN optimization.
The most compelling argument for SD-Branch is operational agility. IT organizations can rapidly deploy and provision a network branch-in-a-box solution for new locations. Via a centralized management console, they can control and adjust all branch network and security functions.
Reducing or eliminating the need for trained IT personnel to visit remote branch locations results in significant cost and time savings. SD-Branch also promises to reduce hardware costs by deploying software on consolidated hardware as compared to many separate appliances.
Other SD-Branch benefits include:
- Decreased cost of support and maintenance contracts because fewer vendors will be involved.
- The ability to right-size hardware requirements for each branch thanks to software virtualization.
- A smaller hardware footprint, which is ideal for space-constrained branches.
- Network performance scalability. As network requirements change, the performance of any function can be tuned up or down by changing processor allocation or adding hardware resources.
- Lower power consumption because one power-efficient platform replaces many appliances.
Over time the SD-Branch will be easier to deploy, less complex to manage, and more responsive to changing requirements at the branch. The cost benefits in CAPEX and OPEX could be significant as the technology matures.
The broader concept of SD-Branch is still in its early stages. During 2018, we will see a number of suppliers introduce their SD-Branch solutions. These initial SD-Branch implementations will primarily be single-vendor and may lack state-of-the-art technology in some applications.
IT leaders should carefully evaluate the benefits of the SD-Branch architecture. Migration to SD-Branch will likely require significant changes to the existing branch network and may require a forklift upgrade. SD-Branch suppliers should be evaluated on their current and near-future technology, technology partnerships (e.g. security), and deployment options (do it yourself, channel partners, and managed solutions).
Brad believes that SD-branch is inevitable. He provided the following rationale for that:
▪SD-Branch will be enabled by SD-WAN overlays
▪ SD-WAN will be integral component of SD-Branch, but latter will include other
virtualized (perhaps containerized) network/security services
▪ Automated provisioning, management, and orchestration results in SD-Branch that
yields dynamic network as a service (NaaS)
• Network and security services added or modified as needed
• Results in CapEx savings (separate hardware appliances no longer need for each network
• Network operating costs are lower, minimized need for branch IT
• Provisioning is far more agile, resulting in faster time to revenue/business outcome
▪ It’s not enough to have virtual network appliances
• Virtual appliances are still appliances architecturally
• SD-Branch gets us to cloud-like NaaS at the branch/remote office
IDC Prediction: Edge IT goes mainstream in 2022, displacing 80% of existing edge appliances.
IDC’s Essential Guidance for Users:
- Consider the role and value of the network not just in terms of connectivity, but in how it can contribute to intelligence at the edge.
- Look for edge-networking solutions that abstract management complexity, provide application-centric automation, speed provisioning, and ensure application availability and security.
- Ensure your intelligent-edge network evolves to a NaaS model, in which virtualized network and security services are dynamically provisioned as needed. The edge network must be as agile as the apps/services it supports.
IDC’s Essential Guidance for Suppliers:
- Continually enhanced intelligent network automation and orchestration to reduce operational complexity and provide network agility.
- Leverage ML/AI as means to the end goal of providing increasingly actionable visibility that loops back to feed intent/policy and allows for proactive remediation.
- Provide for true NaaS at the intelligent edge, incorporating a full range of virtualized network and network-security services (through ecosystem partnerships).