Networking

The thirst for more and more internet speed continues to grow. In a world where the pandemic accelerated digital transformation, one could argue the internet connection into one’s home is critically important to the way we work, learn and play. To date, however, consumers have been limited to gigabit speeds, which might have seemed fast a couple of years ago, but today is putting a cap on the things we can do. AT&T breaks the gig barrier for home internet 

On Monday, AT&T broke the gig barrier when it announced its fiber customers can now get multi-gigabyte (GB) internet speeds, as the carrier doubles down on fiber in its broadband infrastructure. AT&T will offer symmetric 2.5GB and 5GB speed options beginning this week.AT&T is also rolling out simpler pricing for its fiber portfolio without additional equipment fees, annual contracts, and data caps. AT&T Fiber and Business Fiber customers with a 2GB plan will pay $110 per month and $225 per month, respectively. This is ideally suited for small businesses or those who have many connected devices in the home. The 5GB option will cost AT&T Fiber customers $180 per month and Business Fiber customers $395 per month.Symmetric bandwidth can be a game-changer for video users or content creators  The notable parts of the announcements are the symmetric bandwidth, as that’s a rarity with broadband. Comcast Xfinity offers 2GB download speeds, but the upload is limited to 35Mbps, which is a limitation of cable. Verizon offers near symmetric gigabit fiber but not multi-gigabit speeds. Symmetric bandwidth is important for video calls, gaming, and content creators, who are uploading massive files to the cloud. In this case, customers of AT&T would see a marked performance improvement. Also, I’m a big fan of transparent pricing where the cost is fixed in perpetuity. Often, broadband providers offer a low introductory price and then jack the price up after a year. By now, most savvy buyers know that if one calls and complains, they can get the price reduced. Putting customers through this gauntlet annually is one reason why companies like Comcast’s NPS score is so low. This skit by SNL actually parodies a call with Spectrum, which seems like a typical call to your local cable provider. AT&T’s service is no better, but holding the price fixed is at least one less reason for a customer to contact the call center. Also, the price is inclusive of fees, equipment, and other factors that can drive a seemingly low price up. With telecom services, it’s rare that you get what you pay for, but in this case, that’s true. Fiber is a proven technology The fiber network from AT&T is reliable, secure, and tested. It’s used by the U.S. government, the military, first responders, and leading companies with complex connectivity needs. More than 2.75 million U.S. businesses currently rely on AT&T’s high-speed fiber connections.

However, businesses aren’t the only ones with a need for speed. Research cited in the AT&T press material shows the average consumer has 13 connected devices in their home, which could go up to 32 devices or more in the near future. This includes traditional devices, such as tablets and laptops, smart TVs, streaming devices, gaming consoles, appliances, connected doorbells, and more. Such devices consume tons of data and demand more bandwidth.On top of that, more people are working from home due to the COVID-19 pandemic. Multi-gig speeds are primed for these demands and can provide the bandwidth homes and businesses require to run a multitude of connected devices. Fiber was designed specifically for high-speed internet, enabling high-capacity tasks like uploading large files during video calls, as well as gaming and entertainment. AT&T’s multi-gig fiber launch is part of the carrier’s strategy to provide customers with a seamless wireless experience from a single carrier by combining its 5G network and its fiber network. AT&T has also amped up its Wi-Fi technology. Last year, the carrier launched a gateway that’s Wi-Fi 6 and Tri-band enabled to support multiple connected devices.AT&T envisions a future of fiber that’s hyperlocal, hyper-reliable, and hyper-fast. The service will be available in more than 70 metro areas across the country, including Los Angeles, Atlanta, and Dallas, which might seem like a minor number, but it’s currently only available in 5.2 million customer locations, which is a fraction of the country. AT&T will expand this to about 30 million in 2025, which is still the minority of the country. If you’re lucky enough to be in the AT&T footprint, the service is worth a look.

Networking More

Juniper Networks on Tuesday announced a set of new silicon chipsets: The Trio 6 chipset is optimized for a wide range of use cases on the edge, with flexibility to adapt to future networking use cases. Meanwhile, the Express 5 chipset is designed for high throughput, delivering non-blocking throughput of 28.8T in a single package. These new application-specific integrated circuits (ASICs), Juniper says, are designed to be optimized for the specific needs of specific points in the network. “As networks have evolved over the past two decades by supporting more diverse and demanding digital services, operators have increasingly sought out specialized silicon to tackle specific roles,” Juniper’s Brendan Gibbs wrote in a blog post. “Networks run better with ASICs optimized for different tasks.”The sixth generation of Juniper’s Trio silicon for MX Series routers maximizes logical scale and programmability for complex and dynamic edge service nodes. At the network edge, platforms need to be able to support a growing number of diverse business and consumer use cases and features.The Trio 6, which is machine learning-enabled, also helps deliver security with native support for IPSec and integrated MACsec at native line rate. In terms of power usage, the Trio 6 uses 7-nm fabrication technology to deliver a 70% improvement in efficiency compared to previous-generation chipsets.The Trio 6 is available now. Along with the Trio 6, Juniper is rolling out new additions to the MX Series routing portfolio, all based on the new chipset. This includes the Juniper MX10K family, which offers the first 400G-capable LC9600 line card.The new Express 5 ASIC, meanwhile, is designed for PTX10K series platforms. Juniper says it delivers the industry’s  highest non-blocking throughput. Also built with 7-nm technology, it delivers 45% better power efficiency than previous chipsets. 

Express 5 silicon taped out in 2021 and will be available in shipping product at a future date.

Networking More

The race is on to create one neural network that can process multiple kinds of data — a more-general artificial intelligence that doesn’t discriminate about types of data but instead can crunch them all within the same basic structure.

Artificial Intelligence

The genre of multi-modality, as these neural networks are called, is seeing a flurry of activity in which different data, such as image, text, and speech audio, are passed through the same algorithm to produce a score on different tests such as image recognition, natural language understanding, or speech detection. And these ambidextrous networks are racking up scores on benchmark tests of AI. The latest achievement is what’s called “data2vec,” developed by researchers at the AI division of Meta (parent of Facebook, Instagram, and WhatsApp).The point, as Meta researcher Alexei Baevski, Wei-Ning Hsu, Qiantong Xu, Arun Babu, Jiatao Gu, and Michael Auli reveal in a blog post, is to approach something more like the general learning ability that the human mind seems to encompass. “While people appear to learn in a similar way regardless of how they get information — whether they use sight or sound, for example — there are currently big differences in the way self-supervised learning algorithms learn from images, speech, text, and other modalities,” the blog post states.The main point is that “AI should be able to learn to do many different tasks, including those that are entirely unfamiliar.” Meta’s CEO, Mark Zuckerberg, offered a quote about the work and its ties to a future Metaverse: People experience the world through a combination of sight, sound, and words, and systems like this could one day understand the world the way we do. This will all eventually get built into AR glasses with an AI assistant so, for example, it could help you cook dinner, noticing if you miss an ingredient, prompting you to turn down the heat, or more complex tasks.

The name data2vec is a play on the name of a program for language “embedding” developed at Google in 2013 called “word2vec.” That program predicted how words cluster together, and so word2vec is representative of a neural network designed for a specific type of data, in that case text.  Also: Open the pod bay doors, please, HAL: Meta’s AI simulates lip-readingIn the case of data2vec, however, Baevski and colleagues are taking a standard version of what’s called a Transformer, developed by Ashish Vaswani and colleagues at Google in 2017, and extending it to be used for multiple data types. The Transformer neural network was originally developed for language tasks, but it has been widely adapted in the years since for many kinds of data. Baevski et al. show that the Transformer can be used to process multiple kinds of data without being altered, and the trained neural network that results can perform on multiple different tasks.  In the formal paper, “data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Language,” Baevski et al., train the Transformer for image data, speech audio waveforms, and text language representations.  The very general Transformer becomes what is called a pre-training that can then be applied to specific neural networks in order to perform on specific tasks. For example, the authors use data2vec as pre-training to equip what’s called “ViT,” the “vision Transformer,” a neural network specifically designed for vision tasks that was introduced last year by Alexey Dosovitskiy and colleagues at Google. Meta shows top scores for the venerable ImageNet image-recognition competition.
Meta 2022
When used on ViT to try to solve the standard ImageNet test of image recognition, their results come in at the top of the pack, with accuracy of 84.1%. That’s better than the score of 83.2% received by a team at Microsoft that pre-trained ViT lead by Hangbo Bao last year. And the same data2vec Transformer outputs results that are state-of-the-art for speech recognition and that are competitive, if not the best, for natural language learning: Experimental results show data2vec to be effective in all three modalities, setting a new state of the art for ViT-B and ViT-L on ImageNet-1K, improving over the best prior work in speech processing on speech recognition and performing on par to RoBERTa on the GLUE natural language understanding benchmark.  The crux is that this is happening without any modification of the neural network to be about images, and the same for speech and text. Instead, every input type is going into the same network and is completing the same very general task. That task is the same task that Transformer networks always use, known as “masked prediction.”  Also: Google’s Supermodel: DeepMind Perceiver is a step on the road to an AI machine that could process anythingThe way that data2vec performs masked prediction, however, is an approach known as “self-supervised” learning. In a self-supervised setting, a neural network is trained or developed by having to pass through multiple stages. First, the network constructs a representation of the joint probability of data input, be it images or speech or text. Then, a second version of the network has some of those input data items “masked out,” left unrevealed. It has to reconstruct the joint probability that the first version of the network had constructed, which forces it to create increasingly better representations of the data by essentially filling in the blanks. An overview of the data2vec approach.
Meta 2022
The two networks, the one with the full pattern of the joint probability, and the one with the incomplete version that it is trying to complete, are called, sensibly enough, “Teacher” and “Student.” The Student network tries to develop its sense of the data, if you will, by reconstructing what the Teacher has already achieved. You can see the code for the models on Github. How is the neural network performing Teacher and Student for three very different types of data? The key is that the “target” of joint probability in all three data cases is not a specific output data type, as is the case in versions of the Transformer for a specific data type, such as Google’s BERT or OpenAI’s GPT-3. 

Networking

Rather, data2vec is grabbing a bunch of neural network layers that are inside the neural network, somewhere in the middle, that represent the data before it is produced as a final output.  As the researchers write, “One of the main differences of our method […] other than performing masked prediction, is the use of targets which are based on averaging multiple layers from the teacher network.” Specifically, “we regress multiple neural network layer representations instead of just the top layer,” so that “data2vec predicts the latent representations of the input data.” They add, “We generally use the output of the FFN [feed-forward network] prior to the last residual connection in each block as target,” where a “block” is the Transformer equivalent of a neural network layer. The point is that every data type that goes in becomes the same challenge for the Student network of reconstructing something inside the neural network that the Teacher had composed. This averaging is different from other recent approaches to building One Network To Crunch All Data. For example, last summer, Google’s DeepMind unit offered up what it calls “Perceiver,” its own multi-modal version of the Transformer. The training of the Perceiver neural network is the more-standard process of producing an output that is the answer to a labeled, supervised task such as ImageNet. In the self-supervised approach, data2vec isn’t using those labels; it’s just trying to reconstruct the network’s internal representation of the data.  Even more ambitious efforts lie in the wings. Jeff Dean, head of Google’s AI efforts, in October teased about “Pathways,” calling it a “next generation AI architecture” for multi-modal data processing. Mind you, data2vec’s very general approach to a single neural net for multiple modalities still has a lot of information about the different data types. Image, speech, and text are all prepared by pre-processing of the data. In that way, the multi-modal aspect of the network still relies on clues about the data, what the team refer to as “small modality-specific input encoders.” Also: Google unveils ‘Pathways’, a next-gen AI that can be trained to multitaskWe are not yet at a world where a neural net is trained with no sense whatsoever of the input data types. We are also not at a point in time when the neural network can construct one representation that combines all the different data types, so that the neural net is learning things in combination.That fact is made clear from an exchange between ZDNet and the researchers. ZDNet reached out to Baevski and team and asked, “Are the latent representations that serve as targets a combined encoding of all three modalities at any given time step, or are they usually just one of the modalities?” Baevski and team responded that it is the latter case, and their reply is interesting enough to quote at length: The latent variables are not a combined encoding for the three modalities. We train separate models for each modality but the process through which the models learn is identical. This is the main innovation of our project since before there were large differences in how models are trained in different modalities. Neuroscientists also believe that humans learn in similar ways about sounds and the visual world. Our project shows that self-supervised learning can also work the same way for different modalities. Given data2vec’s modality-specific limitations, a neural network that might truly be One Network To Rule Them All remains the technology of the future. More

Increasing connectivity and communication demands are paving the way for private 5G, a cloud-era wireless technology designed for the enterprise and highly adaptable to changes. Many organizations are already implementing or thinking about implementing private 5G because the network and the data can better be controlled by the enterprise. It can also be restricted to a certain location, providing coverage both indoors or outdoors in places such as manufacturing plants and ports.

On top of that, private 5G allows organizations to control and customize their security settings, policies, and other aspects of wireless communications.A new study recently published by Economist Impact in partnership with NTT surveyed organizations around the world and uncovered that more than half of them plan to deploy a private 5G network within the next six to 24 months. The survey included 216 C-level and senior IT decision-makers from organizations with a revenue of $250 mil. to more than $1 billion. The respondents came from various industries in Germany, Japan, the UK, and the U.S., and they included automotive and manufacturing, energy, health care, pharma, retail, and logistics.According to the study, organizations are broadly adopting next-gen connectivity and communications technologies, including private 5G. 94% of the respondents are implementing upgrades that include Wi-Fi 6, 4G, or 5G. Nearly a quarter (24%) are piloting private 5G networks, while 6% have at least one operational private 5G network. Among those with one operational private 5G network, the largest group is from the U.S. (9.3%) followed by Germany (7%), although Germany leads (33%) when it comes to piloting private 5G networks. Energy and transportation lead the way for installing private 5G Private 5G interest is especially high in industrial settings to support smart manufacturing use cases such as robots and self-driving machines. Energy (39%) and transport (33%) are the two industries more likely to be piloting 5G networks. Transport companies (41%) are most likely to have already built a private 5G network. Within the automotive and manufacturing industries, 25% of companies reported having a private 5G pilot and 5& have an operational network. In health care and pharma, 18% of companies are piloting a private 5G network and 5% have an operational network.These industries make sense as network reliability is critical to business operations. Even the smallest hiccup in the wireless network can cost millions of dollars, which is why the verticals listed above have historically stayed away from Wi-Fi, which can be flaky at times. I’m sure everyone reading this has experienced a Wi-Fi network that appears to be working fine and then suddenly stops working and then just as quickly starts again. This is fine in a carpeted office but not on a manufacturing floor. Security is top driver Not surprisingly, security is a key driver for private 5G adoption. 69% of the respondents said network security was not being addressed by their current connectivity and communications platforms, making it a top concern for organizations across countries and industries. For 75% of health care and pharma organizations, security is the biggest pain point, given the sensitive nature of the data. Other key pain points cited by the respondents were control of data (48%), coverage and speed (43%), and the response time of their current service provider (40%). 

Security is the reason why most organizations are exploring solutions beyond Wi-Fi. 87% of the respondents believe Wi-Fi networks don’t provide a sufficient level of security for the enterprise. In fact, most (86%) of the respondents believe private 5G is a substitute for Wi-Fi. That’s because private 5G networks offer several advantages to compliance-driven organizations for customizing security and data protection. The other benefits of implementing private 5G cited by the respondents are improved data privacy (83%), faster connection speeds with lower latency (81%), and increased network reliability for connectivity and communications (80%). Although private 5G adoption seems to be speeding up, it’s still in the early stages for most organizations. Implementing private 5G is either in the short- to medium-term plans for organizations that have yet to pilot or implement such networks. Globally, only 3% of companies plan to deploy private 5G within six months, while 15% plan to implement within 12 months, and 19% within 18 months.Building out private 5G infrastructure comes with some technical challenges that organizations shared in the study. For 44% of the respondents, a major barrier is integrating 5G with legacy systems and networks. Complexity around the infrastructure needed to deploy 5G (37%) and employees lacking technical skills to manage 5G networks (30%) are the other barriers to private 5G adoption. Managed services as a viable option for deployment For this reason, many organizations prefer to outsource their private 5G deployments. 38% of organizations choose to outsource to a managed service provider with service-level agreements; meanwhile, one-third of organizations would rather have a hybrid or shared private network approach, where they lease the network from a mobile operator. When it comes to engaging with private 5G suppliers, organizations are most likely to request system integration services (63%), post-deployment network management (62%), and network design and planning (54%).The study’s findings show adopting private 5G networks is strongly supported by senior leadership across the globe. Looking ahead, 94% of the respondents agree that 5G will become an important part of their operations. More than 90% envision private 5G becoming a standard in their industry within the next five years — a view that is shared across all sectors. It will also be the catalyst for enabling digital transformation in the enterprise.It’s important to understand the positioning of 5G versus Wi-Fi. Some industry watchers have predicted that 5G would eat away at Wi-Fi, but that’s certainly not the case. I believe the two to be highly complementary with Wi-Fi continuing to be the wireless standard of choice for general use cases and 5G when guaranteed, reliable connectivity is needed. A proof point of this comes from this Deloitte study that found that 98% of businesses will use both technologies within three years. More

Comcast revealed that it has successfully tested a new prototype DOCSIS 4.0 modem that is designed to bring 10G technology into customers’ homes for the first time.

According to the broadband provider, the new unit has achieved symmetrical download and upload speeds in excess of 4 gigabits per second (Gbps) thanks to its “Full Duplex DOCSIS 4.0 system-on-chip (SoC).” While these figures were collected in a laboratory environment, Comcast claims the new model is capable of even faster data transmission rates in the future, as the company continues to chase the eponymous 10Gbps potential transfer rates promised by 10G networks. The cable company’s product reveal is just the latest stop on the long road it has been on to make 10G technology viable for consumer broadband. Previous milestones have included testing 10G connections over a virtualized cable modem termination system (vCMTS) using the same DOCSIS 4.0 technology found in the new modem and an earlier test of a 10G SoC, which used Network Function Virtualization (NFV) technology and Comcast’s live residential network to reach a more modest 1.25Gbps. The use of its existing nationwide network is a major goal for Comcast, which touted the fact that DOCSIS 4.0 can allow 10G transmissions via its existing cable infrastructure, with only the modem at endpoints in user homes likely needing to be replaced in most markets. Comcast clearly sees 10G technology as the future of its home broadband offerings, noting that even 4Gbps can be exceeded “as developers refine technology at every level of the 10G architecture.”For comparison, the company’s residential broadband plans currently top out in most areas with its Gigabit tier, which offers 1Gbps to 1.2Gbps download speeds, with some select regions gaining access to its Gigabit Pro service, which rises to 2Gbps. However, these speedy plans currently only support much, much slower upload rates of just 35Mbps. Comcast was previously called out for hiding this fact by Ars Technica, which noted how difficult it is to find an actual upload rate across the company’s various sign-up pages. While download rates tend to be far more important for the average consumer than upload rates, Comcast’s relatively slow upload speeds are something fiber broadband companies have kept as an advantage over it. Many fiber-based plans from companies like Verizon and Google already offer symmetrical rates that reach or come close to 1Gbps both up and down. In addition to the faster download speeds, the symmetrical transfer rates promised by this new modem may be just as important for customers that Comcast has never previously been able to capture with its existing, slower uploads. 

The company did not provide any timeframe for this technology to reach the general public.

Networking More

Every year, Apple discontinues a swathe of products, and 2022 will be no exception.This is a normal process for tech companies, so knowing ahead of time helps for two reasons: you can avoid buying products that will discontinued, and you will be aware if your own devices are headed for end-of-life.See also: Don’t waste your money on these Apple products: January 2022 editionA good way to predict what will be discontinued in 2022 is to look at what was discontinued in 2021. Here’s a list of 2021’s discontinued tech:iPhone 12 Pro and 12 Pro MaxApple Watch Series 6iPhone XRiPad 8th-geniPad mini 5th-gen21.5-inch iMaciMac ProHomePodApple TV 4K 1st-genBeats EP/Solo Pro/Powerbeats 3Some of these were to be expected to make room for new generations (iPhones, iPads, and Apple Watch). It’s logical, but some of the others were interesting — especially the HomePod, 21.5-inch iMac, and the iMac Pro. These point to a shift of priorities at Apple.So, what should we expect Apple to discontinue in 2022?

Here are my predictions:iPhones 13 Pro, 13 Pro Max, and 12: These will be superseded by new iPhones in September.Apple Watch Series 7: Will be superseded by a new Apple Watch.iPhone SE: At almost two years old, this is heading towards the end of the line.iPad Air 4th-gen: By September 2022, this will be two years old and ripe for dropping.iMac 27-inch: This will get the M1 Apple Silicon treatment.Mac Pro: This will also get the M1 Apple Silicon treatment.AirPods Pro: Released in October 2019, these have had a good run; I’m expecting an updated version to land this year.iPod touch 7th-gen: Released in May 2019, this thing is now a complete dinosaur. I fully expect Apple to kill the iPod touch once the current one is discontinued, as it’s hard to see it being a viable product now.As usual, it looks like it’s going to be a big year for Apple updates. Stay tuned. More

Huawei Technologies warns that it will see “serious challenges” in 2022, amidst an uncertain business environment, “politicisation of technology”, and further “deglobalisation”. It also reveals plans to streamline decision-making processes in its local offices next year, giving these outfits more autonomy. The Chinese tech vendor is expecting to close the year with 634 billion yuan ($99.45 billion) in revenue, a 28.88% dip from 891.4 billion yuan in 2020. Its carrier business had stayed “stable” and its enterprise unit saw growth, said Huawei’s rotating chairman Guo Ping, in his new year message Friday to employees. He added that digital transformation in global economies had become a major growth engine and there were new opportunities in green and low-carbon technologies, but warned of uncertainties in the year ahead, 

“An unpredictable business environment, the politicisation of technology, and a growing deglobalisation movement all present serious challenges,” Guo said. “Against this backdrop, we need to stick to our strategy and respond rationally to external forces that are beyond our control.”He noted that Huawei would push ahead with its focus on infrastructure and smart devices, and look to respond more quickly to customer needs with shorter “management chains”.  This meant creating “integrated teams” and “domain-specific subsidiaries”, he said. Specifically, Huawei in 2022 would look to streamline its business decision-making processes by giving more autonomy to local offices. This would see these outfits assuming the authority to make certain decisions previously held by its Shenzhen headquarters. Further tweaks to its organisational structures could see business integration across its local offices worldwide. Huawei has business operations in more than 170 markets, including 14 offices in the Asia-Pacific region outside of China. 

The main objective of its organisation-wide transformation efforts was to enhance operational efficiencies and customer service delivery, Guo said. Elaborating on its product development plans, he noted that Huawei’s software offerings would revolve around EulerOS while its device portfolio would be driven by HarmonyOS. “These two ecosystems will adhere to an open source strategy, allowing all software developers to use them, contribute to them, and benefit from them,” he said. “We will continue to build and contribute to online developer communities as well as brick-and-mortar innovation centres.”He added that Huawei would increase its investment in HarmonyOS and EulerOS, but gave no details on what these entailed. EulerOS is pitched as Huawei’s infrastructure platform that supports both on-premises and cloud computing services. It runs on Huawei’s version of Linux OS. HarmonyOS currently supports more than 220 million Huawei devices and there are more than 100 million devices developed by third-party vendors that currently run on HarmonyOS, according to Huawei. More investment also would be poured into its digital power business, according to Guo. Set up as a business unit in June 2021, Huawei Digital Power Technologies aims to digitalise traditional energy and build products that integrate digital and power electronics capabilities.He also pointed to growth potential in the automotive industry, where he aimed for Huawei to be a “preferred provider” of new components in intelligent vehicles. The Chinese vendor this year would spend $1 billion in research and development (R&D) for intelligent automotive components.US President Joe Biden last month passed a legislation that banned companies such as Huawei and ZTE from getting approval for network equipment licences in the US. The Secure Equipment Act of 2021 would require the Federal Communications Commission (FCC) to adopt new rules stating it would no longer review or approve any authorisation applications for networking equipment that posed national security threats.The FCC in 2020 labelled Huawei and ZTE as national security threats, pointing to both companies’ close ties to the Chinese Communist Party and China’s military.Huawei previously called out the US government’s move to restrict semiconductor exports as another attempt to stem foreign competition. The Chinese vendor had been added to the US government’s Entity List, prohibiting US companies from transferring goods to companies on the list unless they had procured a licence from the US government. The move prompted Huawei to increase its research and development investment by 30% as well as invest in reengineering its products, Guo then said. This had led to redesigns of more than 1,800 boards and rewrites of some 16 million lines of its software codes, with the company looking towards alternative sources for many of its materials.The trade and export bans had led to sluggish earnings in recent years, including dips in profits and smartphone sales, as well as disruptions to Huawei’s supply chain, prompting the vendor to diversify its product focus and chip suppliers.RELATED COVERAGE More

When I started networking, 300 bits per second (BPS) was the best you could do from home and our brand new 802.3 Ethernet gave us a big 10 Megabits per second (MBPS) at the office. I wanted more. Today, with cable gigabit to my home office and 2 Gigabit per second (GBPS) in my office. I still want more. Everyone wants faster, better networking. In 2022, we’ll get what we want. 

First, 5G is finally going to actually make a difference. To date, most of what we’ve been getting is 5G hype. You see 5G has three major different network approaches. Only one of these — millimeter-wave (mmWave) — can give you gigabit speeds. But, and it’s a big but, mmWave, which Verizon calls 5G Ultra-Wideband, has a range that’s better measured in feet instead of yards and it can be blocked by ordinary window glass. In short, you can forget about getting gigabit speeds for all practical purposes.  So, why is Verizon advertising it has 5G everywhere? Because it offers another 5G approach: Dynamic Spectrum Sharing (DSS). Verizon 5G DSS performance is the same as you’ve already been getting from Verizon 4G LTE. In short, it’s a pure advertising gimmick with no real speed advantage. DSS does, however, deliver much better latency than 4G LTE. As more and more 4G phones are replaced with 5G models, DSS-based 5G will start delivering more speed. It will never be gigabit speeds, but it will be better than old-style 4G.However, 4G is also getting a speed boost. Verizon is using a new frequency range, Citizen Band Radio Service (CBRS)–which has nothing to do with the CB radio of truckers and Burt Reynolds movies–can increase 4G speeds up to 800 Megabits per second (Mbps). So, don’t be in too much of a hurry to move to 5G. 4G may work better for you in 2022.Another 5G technology, mid-band, is another story. This one, which today is mostly deployed by T-Mobile, has an average speed of 162 Mbps. Now that doesn’t sound fast, but that’s only because you’ve been drinking 5G marketing kool-aid. It’s the fastest average 5G internet in the US today. The real reason why T-Mobile’s mid-band 5G matters isn’t so much its speed as its range. Instead of feet or yards, it has a range in miles. This makes it ideal for bringing broadband to users who don’t live in cities. If you live in the country, T-Mobile is probably your best choice for your smartphone, office, and home. With 41% 5G coverage of the country, T-Mobile out-distances its 5G rivals.  

Another big advance is that while some of us are lucky enough to live in places with fast broadband, many of us are still stuck with DSL or even–shudder–dial-up internet connections, things are finally changing for the better. Jessica Rosenworcel, a net neutrality proponent, is now the Federal Communications Commission (FCC) Chair. What’s even more important is that Congress passed President Joe Biden’s Infrastructure Investment and Jobs Act. This act included more than $65 billion to build out broadband networks and make broadband more affordable. How this will be spent depends on each state, but the bottom line is it will end up paying to get broadband deployed to places that have never dreamed of seeing real internet speed.Some of that money will be spent on broadband that doesn’t need new, expensive cable buildouts. Instead, it will go to Low-Earth Orbit (LEO) satellites such as SpaceX’s Starlink. Starlink is finally moving from beta to production. While Starlink isn’t as fast as the top cable or fiber networks, it’s still much faster than most rural users have ever seen. For all the buzz Starlink has generated, keep two things in mind. First, Starlink terminal production has gone much slower than originally planned. For example, I’ve been on the waitlist for nine months now and I was recently told I can’t expect to see my unit until some time in the summer of 2022. If you were to order one today, you can expect to wait until the end of 2022 or early 2023. Ouch!Another problem is that because of the demand, Starlink’s average download speed has declined from 97.23 Mbps during Q2 2021 to 87.25 Mbps in Q3 2021. That’s still a lot faster than what my friends and family currently get back in Calhoun County WV and other rural areas, but it’s still a disturbing trend.Finally, if you’re still using 3G, and I know some of you are, get ready for your old phones to stop working. All the major cell companies are turning off their 3G services. AT&T is shutting down its 3G services in February 2022; T-Mobile flips the switch on March 31, 2022; and Verizon will hold off the longest and pull the plug in December 2022.All-in-all, next year should see most of us get faster internet no matter where we live or what services we use. We’ll still want more, of course, we always do, but 2022 will still be a good one for internet users.Related Stories:

ZDNet Recommends More