Aftermath of Salt Typhoon cyberattack: How to secure U.S. telecom networks?
Salt Typhoon Attack: On December 4, 2024, a top U.S. security agency representative confirmed reports that foreign actors, state-sponsored by the People’s Republic of China, infiltrated at least eight U.S. communications companies, compromising sensitive systems and exposing vulnerabilities in critical telecommunications infrastructure. This was part of a massive espionage campaign that has affected dozens of countries. Salt Typhoon has targeted telcos in dozens of countries for upward of two years, officials added.
Dated legacy network equipment and years of mergers and acquisitions are likely impeding the ability of telecommunications providers to prevent China inspired cyber-attacks. Until telecom operators fully secure their networks, China will keep finding ways to come back in, officials have warned.
- On Thursday, FCC chair Jessica Rosenworcel proposed a new annual certification requirement for telecom companies to prove they have an up-to-date cybersecurity risk management plan. More below.
- Senior Cybersecurity and Infrastructure Security Agency and FBI officials confirmed Tuesday that U.S. telcos are still struggling to keep the China-backed hackers out of their networks — and they have no timeline for when total eviction is possible.
FCC Chair Jessica Rosenworcel suggested ‘telecom carriers’ raise their network security methods and procedures: “The cybersecurity of our nation’s communications critical infrastructure is essential to promoting national security, public safety, and economic security,” said Rosenworcel. “As technology continues to advance, so does the capabilities of adversaries, which means the U.S. must adapt and reinforce our defenses. “While the Commission’s counterparts in the intelligence community are determining the scope and impact of the Salt Typhoon attack, we need to put in place a modern framework to help companies secure their networks and better prevent and respond to cyberattacks in the future.”
Rosenworcel’s plan is to make U.S. telcos submit some kind of annual certification to the FCC, proving their cybersecurity measures are up to scratch. The clear inference from the attack itself and all the subsequent attempts to shut the stable door after the horse has bolted is that those efforts currently fall short of the mark. Understandably, none of the specific deficiencies have been publicly detailed.
These proposed FCC measures have been made available to the five members of the Commission. They may choose to vote on them at any moment. If adopted, the Declaratory Ruling would take effect immediately. The Notice of Proposed Rulemaking, if adopted, would open for public comment the cybersecurity compliance framework, which is part of a broader effort to secure the nation’s communications infrastructure.
The FCC press release refers to a recent WSJ report based on an unpublished briefing from U.S. national security adviser Anne Neuberger, in which she detailed the scale of the Salt Typhoon attack. “The Chinese compromised private companies, exploiting vulnerabilities in their systems as part of a global Chinese campaign that’s affected dozens of countries around the world,” she was quoted as saying.
Illustration: Sarah Grillo/Axios
……………………………………………………………………………………………………………………………
Legacy network equipment and years of acquisitions have made it particularly difficult for telcos to patch every access point on their networks, Cliff Steinhauer, director of information security and engagement at the National Cybersecurity Alliance, told Axios.
- Many of the systems in question are nearly 50 years old — like landline systems — and they were “never meant for the type of sensitive data and reliance that we have on them right now,” he said.
- During an acquisition, a company could also miss a server when taking stock of all its newly acquired equipment, Steinhauer said. Network engineers are often inundated with security alerts that are hard to prioritize, he added.
- U.S. telecommunications carriers are required to provide a way for law enforcement to wiretap calls as needed — providing another entry point for adversaries.
Many of the security problems telcos face require simple fixes, like implementing multifactor authentication or maintaining activity logs.
- Even CISA’s recent guidance for securing networks focuses on the security basics.
- But to keep China out, telcos would have to make sure that every device — including their legacy physical equipment, online servers and employees’ computers — is patched.
Most high-profile cyberattacks across industries come down to the basics: a compute server that didn’t have multifactor authentication turned on or an employee who was tricked into sharing their password. Even if a company invests all of its resources in cybersecurity, it may not be enough to fend off a sophisticated nation-state like China.
- These actors are skilled at covering their tracks: They could delete activity logs, pose as legitimate users, and route their traffic through compromised computers in the U.S. so they aren’t detected.
- “You’ve got a persistent, motivated attacker with vast resources to poke and prod until they get in,” Mr. Steinhauer said.
References:
https://docs.fcc.gov/public/attachments/DOC-408015A1.pdf
https://www.axios.com/2024/12/06/telecom-cybersecurity-china-hack-us
WSJ: T-Mobile hacked by cyber-espionage group linked to Chinese Intelligence agency
NTIA and DoD report: Spectrum Sharing strategy for users of lower 37 GHz band
In a new 31-page report, the U.S. NTIA and the DoD offered spectrum sharing recommendations for federal and non-federal users in the lower 37GHz band. The NTIA is the government agency that handles federal usage of spectrum. The FCC handles commercial spectrum usage.
“The FCC, NTIA and DoD began discussions in 2020 on the details of a coordination mechanism. These discussions resulted in a draft sharing framework, based on first-in user rights,” the NTIA wrote in a press release. Here’s the backgrounder from NTIA:
- Building on prior collaborative efforts of NTIA, DoD, and the FCC, the findings reflect coordination across a range of government and industry stakeholders.
- The recommendations for a sharing framework take advantage of the physical characteristics of this band, which is well-suited for short-range and line-of-sight wireless applications.
- The report recognizes the need for flexible access tailored to both Federal and non-Federal user requirements to foster technological advances and policy innovation.
With limited incumbent uses, the band presents a “clean slate” for developing a new model for co-primary Federal and non-Federal access. Specifically, this spectrum supports the creation of very narrow, directed beams and limited propagation for ground communications, making robust forms of sharing possible.
- U.S. policymakers have long recognized the unique sharing opportunities of the Lower 37 GHz band, as well as the need to protect Federal sites, including 15 military sites, five National Aeronautics and Space Administration receiving earth station operations and two National Science Foundation sites.
- In coordination with NTIA, the FCC in 2016 adopted an Order that concluded that non-Federal fixed and mobile applications can share 37-38.6 GHz with DoD operations. The Order made the Lower 37 GHz band available for co-primary sharing, with both Federal and non-Federal users accessing the band by registering sites through a coordination mechanism.
- In 2019, the FCC established service rules addressing Federal sites for a 2020 auction in the 37 GHz band, with sharing rules for 37.0-37.6 GHz to be addressed at a later date. Among other things, the decision added one Federal site to the list of protected Federal sites in the 37 GHz band and limited future DoD access to the 37.6-38.6 GHz (Upper 37 GHz) band unless the Department could demonstrate that its operations cannot be accommodated in the Lower 37 GHz band.
- To enable an innovative sharing approach for the Lower 37 GHz band, the FCC, NTIA and DoD began discussions in 2020 on the details of a coordination mechanism. These discussions resulted in a draft sharing framework, based on first-in user rights.
- Following the release of the NSS Implementation Plan, the FCC released a 2024 Public Notice, that sought information on sharing issues in the Lower 37 GHz band, including how to accommodate various use cases through a coordination mechanism between Federal and non-Federal operators.
Image Credit: Dmytro Razinkov/Alamy Stock Photo
The NTIA outlined a detailed, two-phase sharing process for the 37GHz band: “The first phase would use simple propagation models to determine whether there are overlapping contours and permit operations to proceed in the absence of any overlap,” the NTIA wrote. “The second phase would apply in the event of overlap between a proposed site registration and an existing site already registered in the database and would require the parties to exchange more detailed data and attempt to coordinate their operations.” Next the FCC would have to implement the NTIA’s sharing recommendations.
Expected uses of the Lower 37 GHz band include data-intensive applications, such as high speed, low latency 5G services. Wireless operators view this spectrum as well-suited for providing additional bandwidth, for example during large events through indoor distributed antenna deployments. Industry also sees value in the band for addressing increased demand for mobile network capacity by offloading traffic from other bands.
- Potential use cases include fixed wireless access; high-capacity backhaul; cable supplement for Internet of things (IoT) networks and augmented reality applications; and mobile or private networks that support industrial IoT, smart factories and other high-bandwidth indoor communications applications.
- Federal users, including DoD, may leverage some of this same technology, including as part of potential additional adaptations to meet mission requirements (e.g., hardening).
- Although not being proposed for any specific frequency allocation at this time, DoD is evaluating additional use cases to meet military missions, including: (1) Unmanned Systems to provide terrestrial or maritime to aeronautical mobile and potentially space to aeronautical mobile (maritime, terrestrial) unmanned systems; and (2) Wireless Power Transfer to provide a variety of capabilities currently in development by military research labs to deliver power to wireless communication systems, mobile vehicles, surface and subsurface vehicles, and other potential uses cases.
References:
https://www.lightreading.com/5g/dod-agrees-to-spectrum-sharing-paradigm-in-37ghz
T-Mobile and Charter propose 5G spectrum sharing in 42GHz band
Big 5G Conference: 6G spectrum sharing should learn from CBRS experiences
mmWave Coalition on the need for very high frequency spectrum; DSA on dynamic spectrum sharing in response to NSF RFI
Quantum Computers and Qubits: IDTechEx report; Alice & Bob whitepaper & roadmap
Introduction:
In the last decade, the number of companies actively developing quantum computer hardware has quadrupled. Between 2022 and 2024 multiple funding rounds surpassing US$100 million have been closed, and the transition from lab-based toys to commercial product has begun. Competition is building in the quantum computing market, not only between different companies but between quantum computing technologies. The focus today has intensified on the need for logical or error-corrected qubits [1.]. The challenge ahead is to scale up hardware and increase qubit number while reducing errors as well as infrastructure demand. Leaders today have between 1 and 50 logical qubits, thousands are likely needed to provide a meaningful advantage over classical computing alternatives.
Note 1. Quantum computing is based on the use of qubits – the quantum equivalent to classical bits – the architectures available to create them vary substantially. Many are now familiar with IBM and their superconducting qubits – housed inside large cryostats and cooled to temperatures colder than deep space. Indeed, in 2023 superconducting quantum computers broke the 1000 qubit milestone – with smaller systems made accessible via the cloud for companies to trial out their problems.
However, many agree that the highest value problems – such as drug discovery – need many more qubits, perhaps millions more. As such, alternatives to the superconducting design, many proposing more inherent scalability, have received investment. There are now more than eight technology approaches meaningfully competing to reach the million-qubit milestone.
The quantum computing hardware market today has the unique property of seeing rapid growth in revenue generation despite remaining at a low technology readiness level. National laboratories and supercomputing centers are already investing in the installation of early-stage machines on premises, primarily for research but also to allow more local users the ability to ‘pay to play’. This is, in part, a result of the intensifying governmental stake in the technology – and its potential to provide significant economic and national security advantages in conjunction with quantum sensing and quantum communications. As a result, while multiple technical challenges remain, it appears that the race to commercial advantage could well be paved with gold for some. However, towards the end of the decade, as pressure mounts to deliver commercial value and return on investment – some of those leading the charge today may not necessarily prove to be the true winners in the long term.
With so many competing quantum computing technologies across a fragmented landscape, determining which approaches are likely to dominate is essential in identifying opportunities within this exciting industry. IDTechEx uses an in-house framework for quantum commercial readiness level to measure how quantum computer hardware is progressing in comparison with its classical predecessor. Furthermore, as the initial hype around quantum computing begins to cool investors will increasingly demand demonstration of practical benefits, such as quantum supremacy for commercially relevant algorithms. As such, hardware developers need to show not only the quality and quantity of qubits but the entire initialization, manipulation, and readout systems. Improving manufacturing scalability and reducing cooling requirements are also important, which will create opportunities for methodology agnostic providers of infrastructure such as speciality materials and cooling systems. By evaluating both the sector and competing quantum computing technologies, this report provides insight into the opportunities provided by this potentially transformative technology.
……………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………..
Alice & Bob, a leading innovator in fault-tolerant quantum computing, just released their whitepaper and technology roadmap titled, “Think Inside the Box: Quantum Computing with Cat Qubits.”
Key highlights of the whitepaper:
- Exponential Error Reduction: Cat qubits simplify error correction by reducing it from a 2D to a 1D problem, achieving unmatched fidelity (99.999999%) and reducing hardware requirements by up to 200x compared to traditional approaches.
- Roadmap Milestones: Alice & Bob’s plan moves from mastering single qubits to developing commercially viable quantum computers by 2030, with transformative applications across industries such as finance, healthcare, and cybersecurity.
- Quantum Advantage: Their technology positions them to deliver practical solutions to computational problems that are currently beyond the reach of classical computing.
Image Credit: Alice & Bob
………………………………………………………………………………………………………………………………………………………………………………………………
The roadmap details five key milestones in Alice & Bob’s plan to deliver a universal, fault-tolerant quantum computer by 2030:
- Milestone 1: Master the Cat Qubit Achieved in 2024 with the Boson chip series, this milestone established a reliable, reproducible cat qubit capable of storing quantum information while resisting bit-flip errors. Milestone 2: Build a Logical Qubit Currently under development with the Helium chip series, this stage focuses on creating the company’s first error-corrected logical qubit operating below the error-correction threshold.
- Milestone 3: Fault-Tolerant Quantum Computing With the upcoming Lithium chip series, Alice & Bob aims to scale multi-logical-qubit systems and demonstrate the first error-corrected logical gate.
- Milestone 4: Universal Quantum Computing The Beryllium chip series will enable a universal set of logical gates enabled by magic state factories and live error correction, unlocking the ability to run any quantum algorithm.
- Milestone 5: Useful Quantum Computing The Graphene chip series, featuring 100 high-fidelity logical qubits, will deliver a quantum computer capable of demonstrating quantum advantage in early industrial use cases by 2030, integrating into existing high-performance computing (HPC) facilities.
“Our roadmap lays out a clear path to solving quantum’s toughest engineering challenges,” said Raphael Lescanne, CTO and Co-Founder of Alice & Bob. “Quantum computing can seem opaque, but it shouldn’t be. This white paper makes our technology and roadmap accessible for engineers, business leaders and tech enthusiasts alike.”
Achieving practical quantum advantage requires overcoming the errors inherent in quantum systems. Quantum error correction typically relies on additional qubits to detect and correct these errors, but the resource requirements grow quadratically with complexity, making large-scale, useful quantum computing a significant challenge.
Alice & Bob’s cat qubits offer a promising solution to this bottleneck. These superconducting chips feature an active stabilization mechanism that effectively shields the qubits from some external errors. This unique approach has enabled cat qubits to set the world record for bit-flip protection, one of the two major types of errors in quantum computing, effectively eliminating them.
This protection reduces error correction from a 2D problem to a simpler, 1D problem, enabling error correction to scale more efficiently. As a result, Alice & Bob can produce high-quality logical qubits with 99.9999% fidelity, what they call a “6-nines” logical qubit, using a fraction of the resources required by other approaches.
“Quantum computing should be a tool for solving useful problems in science and industry. This white paper shows how Alice & Bob’s cat qubits can bring that vision to life in a practical way by the decade’s end,” said Théau Peronnin CEO and co-founder of Alice & Bob.
References:
https://alice-bob.com/products/solution-the-box/
Bloomberg on Quantum Computing: appeal, who’s building them, how does it work?
China Mobile verifies optimized 5G algorithm based on universal quantum computer
Can Quantum Technologies Crack RSA Encryption as China Researchers Claim?
Quantum Technologies Update: U.S. vs China now and in the future
AT&T will be “quantum ready” by the year 2025
AT&T to deploy Fujitsu and Mavenir radio’s in crowded urban areas
AT&T announced today that it has signed new agreements with Fujitsu and Mavenir to develop radios specifically for crowded urban areas in its Open RAN deployment using Ericsson hardware and software. The goal is to improve network performance and coverage in cities with lots of mobile data traffic.
These radios will be open C-band radios (TDD 4T4R) and dual band radios (B25/B66 FDD 4T4R) which can be attached to existing utility and light poles. They can often be hidden, making them virtually unseen from street level. We are continuing to look for opportunities to bring additional third-party radios into the network when needed.
All open radios will be managed by Ericsson’s Intelligent Automation Platform (EIAP) via open management interfaces. EIAP is Ericsson’s open network management and service orchestration platform. It supports replacing the old legacy equipment and installing the new radios without missing a beat.
When Open RAN architectures are combined with innovative applications called ‘rApps’ from either the operator or third parties, they can greatly improve the customer experience. This is achieved through better network performance, wider coverage, cost efficiency, and fosters innovation. ‘rAPPs’ are expected to play a critical role in managing and sustaining third party radio innovation opportunities.
AT&T is moving 70% of its 5G network traffic to flow across Open RAN hardware by late 2026 – our customers can relax and enjoy a better wireless experience.
…………………………………………………………………………………………………..
Mavenir has been selling open RAN software for years, but it entered the 5G radio sector in 2022 with its OpenBeam brand. Mavenir’s radios for AT&T will be managed by Ericsson’s Intelligent Automation Platform (EIAP).
AT&T said it would only use Mavenir radios in “crowded urban areas,” which are typically covered by small cell radios rather than massive macro cell sites. The operator did not say how many Mavenir radios it would use nor when it might start deploying those radios.
“Maybe the initial thinking is it’s small cells, but there’s a bigger strategy at play here,” AT&T’s Jeff McElfresh said during a media event on Tuesday. McElfresh explained that small cells could play an important role inside AT&T’s network as network traffic increases. After all, small cells are viewed as a way to increase overall wireless network capacity in the absence of additional spectrum.
Mavenir’s other 5G radio customers include Paradise Mobile and Triangle Communications.
Aramco Digital, the tech-focused subsidiary of oil giant Saudi Aramco, is poised to invest $1 billion into Mavenir for a significant minority stake in the business.
That cash is needed. S&P Global recently warned that Mavenir is close to default or restructuring because it has insufficient funds to cover looming debt obligations.
References:
https://about.att.com/blogs/2024/open-ran-new-collaborations.html
https://www.lightreading.com/open-ran/at-t-to-deploy-radios-from-mavenir
NTT advert in WSJ: Why O-RAN Will Change Everything; AT&T selects Ericsson for its O-RAN
New FCC Chairman Carr Seen Clarifying Space Rules and Streamlining Approvals Process
Incoming Federal Communications Commission (FCC) Chairman Brendan Carr will place a major focus on deregulation of commercial space activities and streamlining the approvals processes, space policy according to Communications Daily. Carr has said faster licensing and permitting of commercial space operations will be a priority.
FCC Commissioner Brendan Carr is the senior Republican on the FCC, having served previously as the FCC’s General Counsel. Nominated by both President Trump and President Biden, Carr has been confirmed unanimously by the Senate three times. Described by Axios as “the FCC’s 5G crusader,” Carr has led the FCC’s work to modernize its infrastructure rules and accelerate the buildout of high-speed networks. His reforms cut billions of dollars in red tape, enabled the private sector to construct high-speed networks in communities across the country, and extended America’s global leadership in 5G.
FCC Commissioner Brendan Carr has been nominated by Donald Trump to be the next FCC Chairman. Photo Credit: FCC
In addition, some expect long-awaited clarity on what agency oversees novel space missions like in-orbit servicing, assembly and manufacturing, or asteroid mining. Moreover, the experts anticipate increased openness about the use of satellite communications in federal programs fighting the digital divide.
Satellite Industry Association (SIA) President Tom Stroup said a policy shift that would include satellites becoming a qualified provider of broadband for federally funded programs is likely. The Biden administration effectively excluded satellite use, and there needs to be “an honest discussion” about the sector’s potential role, space lawyer Jim Dunstan said.
Space policy is often relatively consistent from administration to administration, though with some tweaks, said Michelle Hanlon, executive director of the Center for Air and Space Law at the University of Mississippi School of Law. She said that aside from a reduction of regulatory red tape, there also could be a reconciling of agencies’ orbital debris policies. Commercial space operators generally would like to see Commerce’s Office of Space Commerce as the regulator of novel space activities, as long as it is adequately staffed. But regardless of who has the oversight, the biggest need is a final decision on authority, she said. The Republican trifecta of the White House, the House and Senate could bring a resolution, Hanlon said.
The FCC under Carr might not be as aggressive as it has been on orbital debris regulation, but it won’t neglect it altogether, emailed Michael Dodge, University of North Dakota space studies associate professor. He said the space industry likely sees Carr as friendly toward its goals.
Summit Ridge Group’s Armand Musey stated that it’s hard to overstate SpaceX’s impact on the commercial space sector, and its progress will likely dominate most investment decisions. He said Carr’s seeming openness toward allowing satellite participation in broadband access programs like BEAD and the rural digital opportunity fund will benefit Starlink, but also Viasat and the industry broadly. At SpaceX’s urging, Carr is also likely more open than Democrats to increases in power levels for non-geostationary orbit satellite systems, Musey predicted. That could benefit direct-to-device operators, including Starlink, but also AST SpaceMobile, Globalstar and Lynk, he said. M&As under Carr and a Republican DOJ could be somewhat looser for space-related deals, and larger transactions might face fewer conditions, he added.
SpaceX CEO Elon Musk’s friendly relationship with President-elect Donald Trump and Carr could benefit commercial space operators broadly, but it’s also raising eyebrows, sources agree. Musk’s role as co-head of Trump’s Department of Government Efficiency advisory committee will provide him with considerable influence, “and people will wonder where the lines might blur between a business seeking governmental approvals, and an advisor seeking benefits for his business,” Dodge said.
References:
https://www.fcc.gov/about/leadership/brendan-carr
https://www.cnn.com/2024/11/18/media/brendan-carr-trump-fcc-nominee-project-2025/index.html
FCC: More competition for Starlink; freeing up spectrum for satellite broadband service
FCC approves EchoStar/Dish request to extend timeline for its 5G buildout
FCC restores net neutrality order, but court challenges loom large
Analysis: FCC attempt to restore Net Neutrality & U.S. standards for broadband reliability, security, and consumer protection
FCC Draft Net Neutrality Order reclassifies broadband access; leaves 5G network slicing unresolved
FCC increases broadband speed benchmark (x-satellites) to 100/20 Mbit/s
FCC legal advisor: Potential End of ACP Is the ‘Biggest Challenge’ Facing the Broadband Marketplace
Highlights of FCC Notice of Inquiry (NOI) on radio spectrum usage & how AI might be used
TechCrunch: Meta to build $10 billion Subsea Cable to manage its global data traffic
Meta, the parent company of Facebook, Instagram, and WhatsApp, is reportedly planning to build its first fully owned, large-scale fiber-optic subsea cable extending around the globe. The project, spanning over 40,000 kilometers, is expected to require an investment of more than $10 billion. The purpose is to enhance Meta’s infrastructure to meet the growing demand for data usage driven by its artificial intelligence (AI) products and services, according to a report by TechCrunch. Meta accounts for 10% of all fixed and 22% of all mobile traffic and its AI investments promise to boost that usage even further.
TechCrunch has confirmed with sources close to the company that Meta plans to build a new, major, fiber-optic subsea cable extending around the world — a 40,000+ kilometer project that could total more than $10 billion of investment. Critically, Meta will be the sole owner and user of this subsea cable — a first for the company and thus representing a milestone for its infrastructure efforts.
Image Credit: Sunil Tagare
………………………………………………………………………………………………………………………………………….
Meta’s infrastructure work is overseen by Santosh Janardhan, who is the company’s head of global infrastructure and co-head of engineering. The company has teams globally who look at and plan out its infrastructure — and it has achieved some significant industry figures work for it in the past. In the case of this upcoming project, it is being conceived out of the company’s South Africa operation, according to sources.
Fiber-optic subsea cables have been a part of communications infrastructure for the last 40 years. What’s significant here is who is putting the money down to build and own it — and for what purposes.
Meta’s plans underscore how investment and ownership of subsea networks has shifted in recent years from consortiums involving telecoms carriers, to now also include big tech giants. According to telecom analysts Telegeography, Meta is part-owner of 16 existing networks, including most recently the 2Africa cable that encircles the continent (others in that project are carriers including Orange, Vodafone, China Mobile, Bayobab/MTN and more). However, this new cable project would be the first wholly owned by Meta itself.
That would put Meta into the same category as Google, which has involvement in some 33 different routes, including a few regional efforts in which it is the sole owner, per Telegeography’s tracking. Other big tech companies that are either part owners or capacity buyers in subsea cables include Amazon and Microsoft (neither of which are whole-owners of any route themselves).
There are a number of reasons why building subsea cables would appeal to big tech companies like Meta.
First, sole ownership of the route and cable would give Meta first dibs in capacity to support traffic on its own properties.
Meta, like Google, also plays up the lift it has provided to regions by way of its subsea investments, claiming that projects like Marea in Europe and others in Southeast Asia have contributed more than “half a trillion dollars” to economies in those areas.
Yet there is a more pragmatic impetus for these investments: tech companies — rather than telecoms carriers, traditional builders and owners of these cables — want to have more direct ownership of the pipes needed to deliver content, advertising and more to users around the world.
According to its earnings reports, Meta makes more money outside of North America than in its home market itself. Having priority on dedicated subsea cabling can help ensure quality of service on that traffic. (Note: this is just to ensure long-haul traffic: the company still has to negotiate with carriers within countries and in ‘last-mile’ delivery to users’ devices, which can have its challenges.)
References:
Meta plans to build a $10B subsea cable spanning the world, sources say
Google’s Bosun subsea cable to link Darwin, Australia to Christmas Island in the Indian Ocean
“SMART” undersea cable to connect New Caledonia and Vanuatu in the southwest Pacific Ocean
Telstra International partners with: Trans Pacific Networks to build Echo cable; Google and APTelecom for central Pacific Connect cables
Orange Deploys Infinera’s GX Series to Power AMITIE Subsea Cable
NEC completes Patara-2 subsea cable system in Indonesia
SEACOM telecom services now on Equiano subsea cable surrounding Africa
Google’s Equiano subsea cable lands in Namibia en route to Cape Town, South Africa
China seeks to control Asian subsea cable systems; SJC2 delayed, Apricot and Echo avoid South China Sea
HGC Global Communications, DE-CIX & Intelsat perspectives on damaged Red Sea internet cables
Technavio: Silicon Photonics market estimated to grow at ~25% CAGR from 2024-2028
The global silicon photonics market size is estimated to increase by $5.24 billion from 2024-2028, according to Technavio. The market is estimated to grow at a CAGR of almost 24.88% during the forecast period. Increasing need for higher bandwidth is driving market growth, with a trend towards emergence of optical data centers. However, lack of global standards and guidelines poses a challenge.
The decades old global Silicon Photonics market is now experiencing significant growth due to the increasing demand for high-speed data transfer in various industries. With Internet traffic from cloud computing, 5G technology, IoT, and AI-powered devices, there is a need for more efficient and low-power solutions.
Silicon photonics uses integrated circuits (ICs) for optical communications, thereby reducing power consumption compared to electronic technologies.
Key components of silicon photonics include transceivers, optical interconnects, lasers, modulators, and photodetectors. These are used in data centers, telecommunication networks, and interconnection networks. The market is also driven by the adoption of 5G network, self-driving cars, and high-speed kits for point-of-care testing and imaging data. Silicon photonics uses optical waveguides, optical modulators, and photodetectors made from silicon, silicon nitride, and other photonic components. These components are more compact and less susceptible to thermal effects compared to traditional fiber-optic solutions. Additionally, the use of high-powered laser sources, thermal stress management, and liquid-crystal cladding helps mitigate thermal effects and improve performance.
The market for silicon photonics is expected to grow in IT & telecommunications and consumer electronics sectors, with applications in broadband services, telecom service providers, and broadband connections. This growth is driven by the need for high-speed data transfer and low power consumption, making silicon photonics a promising solution for the future of optical communications and data storage systems.
Silicon photonics is now targeted at optical data centers, providing enhanced capabilities for data transmission, processing, and storage. By integrating high-speed, high-bandwidth optical interconnects directly onto silicon chips, silicon photonics enables seamless communication between different data center components. This results in faster data transfer rates, lower latency, and increased scalability, making it an ideal solution for modern applications like cloud computing, artificial intelligence, and big data analytics. The continuous growth in cloud-based applications and big data analytics has significantly expanded the scale of data center networks. Silicon photonics, with its advantages over traditional copper-based interconnects, is a crucial technology in addressing the demands for faster and more efficient data center infrastructure.
The communications industry’s growth is driving the demand for silicon photonics due to its ability to transmit wider bandwidth signals with low latency and maintain signal quality during long-distance communication with minimal loss. Silicon photonics is a key technology in optical communication systems, enabling the transfer of large amounts of data at high speeds. Increased bandwidth and low latency requirements have fueled the demand for silicon photonics-based devices such as receivers, transmitters, and modulators.
Communications Industry – Market size and forecast 2018 – 2028 (USD Mn):
The global silicon photonics market is experiencing significant growth due to increasing demand for high-speed data transfer in various industries. With the Internet traffic from cloud computing, 5G technology, IoT, and AI-powered devices, there is a need for more efficient and low-power optical communications solutions.
The proliferation of data centers and cloud computing infrastructure is a major factor driving the market’s growth. Vendors like Cisco Systems Inc. And Intel Corp. Offer silicon photonics solutions for high-speed data transmission in data center environments. The evolution of 5G networks is another significant factor, as silicon photonics supports 5G networks with low latency and high capacity at a low cost and power per bit. With the increasing investment in 5G networks, the demand for silicon photonics is also expected to rise, boosting the growth of the global silicon photonics market through the communications segment.
Silicon photonics vendors include: AIO Core Co. Ltd., ams OSRAM AG, Broadcom Inc., Corning Inc., Hamamatsu Photonics KK, II VI Inc., Infinera Corp., Innolume GmbH, Intel Corp., International Business Machines Corp., IPG Photonics Corp., MACOM Technology Solutions Inc., NKT Photonics AS, Nokia Corp., NVIDIA Corp., OpenLight Photonics Inc., OSCPS Motion Sensing Inc, RANVOUS Inc., Sicoya GmbH, TRUMPF SE Co. KG, and Cisco Systems Inc.
Market Challenges:
- Lack of global standards and guidelines
- Availability of substitute technologies for silicon photonics
- High heat generation by photonic components
Silicon photonics offers a promising solution with its integration of photonics and electronic components on a single silicon chip. However, challenges such as thermal effects, power consumption, and thermal stress in high-powered laser sources remain. Transceivers, optical interconnects, and lasers are key components in this market, along with modulators, photodetectors, and optical waveguides. Data centers, telecommunication, and IT and telecommunications are major end-users, with consumer electronics and automotive industries also adopting silicon photonics for high-speed kits in self-driving cars and point-of-care testing. Optical network infrastructure, including fiber-optic and active optical cables, is a significant application area. The market is expected to grow further with advancements in silicon nitride, optical multiplexers, attenuators, and other photonic components.
The absence of standardized protocols and specifications in the silicon photonics market poses challenges for both manufacturers and customers. Without universally accepted standards, the integration of silicon photonics components into existing optical communication systems and networks becomes complicated. Compatibility issues arise, product development and manufacturing processes are complicated, and implementation costs increase. Furthermore, the lack of clear standards results in inconsistent performance metrics, making it difficult for customers to compare and evaluate different silicon photonics solutions effectively. Standardization is crucial for the widespread adoption of new technologies, and its absence in the silicon photonics market hinders its growth and potential impact on the optical communication industry.
References:
LightCounting: Silicon Photonics chip market to hit $3 billion in 2029
Light Counting on Silicon Photonics and Optical Switching at SC22
Synopsys and Juniper Networks form new company to pursue “open” silicon photonics platform
LightCounting: Q1 2024 Optical Network Equipment market split between telecoms (-) and hyperscalers (+)
https://viodi.com/2013/11/08/silicon-photonics-cisco-intel-see-light-at-the-end-of-the-tunnel/
Latest Ericsson Mobility Report talks up 5G SA networks (?) and FWA (!)
Ericsson’s November 2024 Mobility Report predicts that global 5G standalone (SA) connections will top 3.6 billion by 2030. That compares to 890 million at the end of 2023. Over that same period of time, 5G SA as a proportion of global mobile subscriptions is expected to increase from 10.5% to 38.4%, while average monthly smartphone data consumption will grow to 40 GB from 17.2 GB. By the end of the decade, 80% of total mobile data traffic will be carried by 5G networks.
That rosy forecast is in sharp contrast to the extremely slow and disappointing pace of 5G SA deployments to date. In January, Dell’Oro counted only 12 new 5G SA deployments in 2023, compared to the 18 in 2022. “The biggest surprise for 2023 was the lack of 5G SA deployments by AT&T, Verizon, British Telecom EE, Deutsche Telekom, and other Mobile Network Operators (MNOs) around the globe. As we’ve stated for years, 5G SA is required to realize 5G features like security, network slicing, and MEC to name a few.”
Fifty 5G Standalone enhanced Mobile Broadband (eMBB) networks commercially deployed (2020 – 2023):
The report states, “Although 5G population coverage is growing worldwide, 5G mid-band is only deployed in around 30% of all sites globally outside of mainland China. Further densification is required to harness the full potential of 5G.” Among the report highlights:
- Global 5G subscriptions will reach around 6.3 billion in 2030, equaling 67% of total mobile subscriptions.
- 5G subscriptions will overtake 4G subs in 2027.
- 5G is expected to carry 80% of total mobile data traffic by the end of 2030.
- 5G SA subscriptions are projected to reach around 3.6 billion in 2030.
Source: Ericsson Mobility Report -Nov 2024
“Service differentiation and performance-based opportunities are crucial as our industry evolves,” said Fredrik Jejdling, EVP and head of Ericsson’s networks division. “The shift towards high-performing programmable networks, enabled by openness and cloud, will empower service providers to offer and charge for services based on the value delivered, not merely data volume,” he added.
The Mobility Report provides two case studies in T-Mobile US and Finland’s Elisa – both of which have rolled out network slicing on their 5G SA networks and co-authored that section of the report:
- T-Mobile has been testing a high priority network slice to carry mission-critical data during special events.
- Elisa has configured a slice to support stable, high-capacity throughput for users of its premium fixed-wireless access (FWA) service, called Omakaista.
The Mobility Report doesn’t say if those two telcos are deriving any monetary benefit from network slicing, or more broadly from their 5G SA networks.
……………………………………………………………………………………………………………………………………………………………………………………………………………..
The Fixed Wireless Access (FWA) market has momentum:
- Ericsson predicts FWA connections will reach 159 million this year, up from 131 million in 2023.
- By 2030, connections are expected to hit 350 million, with 80% carried by 5G networks.
- In four out of six regions, 83% or more wireless telcos now offer FWA.
- The number of FWA service providers offering speed-based tariff plans – with downlink and uplink data parameters similar to cable or fiber offerings – has increased from 30% to 43% in the last year alone.
- An updated Ericsson study of retail packages offered by mobile service providers reveals that 79% have a FWA offering.
- There are 131 service providers offering FWA services over 5G, representing 54 percent of all FWA service providers.
- In the past 12 months, Europe has accounted for 73%of all new 5G FWA launches globally.
- Currently, 94% of service providers in the Gulf Cooperation Council region offer 5G FWA services.
- In the U.S. two service providers (T-Mobile US and Verizon) originally set a goal to achieve a combined 11–13 million 5G FWA connections by 2025. After reaching this target ahead of schedule, they have now revised their goal to 20–21 million connections by 2028.
- The market in India is rapidly accelerating, with 5G FWA connections reaching nearly 3 million in just over a year since launch. • An increasing number of service providers are launching FWA based on 5G standalone (SA).
References:
https://www.ericsson.com/en/reports-and-papers/mobility-report/reports/november-2024
https://www.ericsson.com/4ad0df/assets/local/reports-papers/mobility-report/documents/2024/ericsson-mobility-report-november-2024.pdf
5G Advanced offers opportunities for new revenue streams; 3GPP specs for 5G FWA?
FWA a bright spot in otherwise gloomy Internet access market
Where Have You Gone 5G? Midband spectrum, FWA, 2024 decline in CAPEX and RAN revenue
GSA: More 5G SA devices, but commercial 5G SA deployments lag
Vodafone UK report touts benefits of 5G SA for Small Biz; cover for proposed merger with Three UK?
Building and Operating a Cloud Native 5G SA Core Network
Google’s Bosun subsea cable to link Darwin, Australia to Christmas Island in the Indian Ocean
“Vocus is thrilled to have the opportunity to deepen our strategic network partnership with Google, and to play a part in establishing critical digital infrastructure for our region. Australia Connect will bolster our nation’s strategic position as a vital gateway between Asia and the United States by connecting key nodes located in Australia’s East, West, and North to global digital markets,” said Jarrod Nink, Interim Chief Executive Officer, Vocus.
“The combination of the new Australia Connect subsea cables with Vocus’ existing terrestrial route between Darwin and Brisbane, will create a low latency, secure, and stable network architecture. It will also establish Australia’s largest and most diverse domestic inter-capital network, with unparalleled reach and protection across terrestrial and subsea paths.
“By partnering with Google, we are ensuring that Vocus customers have access to high capacity, trusted and protected digital infrastructure linking Australia to the Asia Pacific and to the USA. “The new subsea paths, combined with Vocus’ existing land-based infrastructure, will provide unprecedented levels of diversity, capacity and reliability for Google, our customers and partners,” Nink said.
“Australia Connect advances Google’s mission to make the world’s information universally accessible and useful. We’re excited to collaborate with Vocus to build out the reach, reliability, and resiliency of internet access in Australia and across the Indo-Pacific region,” said Brian Quigley, VP, Global Network Infrastructure, Google Cloud.
Perth, Darwin, and Brisbane are key beneficiaries of this investment and are now emerging as key nodes on the global internet utilizing the competitive and diverse subsea and terrestrial infrastructure established by the Vocus network. Vocus will be in a position to supply an initial 20-30Tbps of capacity per fiber pair on the announced systems, depending on the length of the segment.
References:
Google’s Equiano subsea cable lands in Namibia en route to Cape Town, South Africa
Google’s Topaz subsea cable to link Canada and Japan
“SMART” undersea cable to connect New Caledonia and Vanuatu in the southwest Pacific Ocean
Telstra International partners with: Trans Pacific Networks to build Echo cable; Google and APTelecom for central Pacific Connect cables
HGC Global Communications, DE-CIX & Intelsat perspectives on damaged Red Sea internet cables
Orange Deploys Infinera’s GX Series to Power AMITIE Subsea Cable
NEC completes Patara-2 subsea cable system in Indonesia
SEACOM telecom services now on Equiano subsea cable surrounding Africa
Bharti Airtel and Meta extend 2Africa Pearls subsea cable system to India
China seeks to control Asian subsea cable systems; SJC2 delayed, Apricot and Echo avoid South China Sea
Intentional or Accident: Russian fiber optic cable cut (1 of 3) by Chinese container ship under Baltic Sea
Altice Portugal MEO signs landing party agreement for Medusa subsea cable in Lisbon
2Africa subsea cable system adds 4 new branches
Echo and Bifrost: Facebook’s new subsea cables between Asia-Pacific and North America
Equinix and Vodafone to Build Digital Subsea Cable Hub in Genoa, Italy
Superclusters of Nvidia GPU/AI chips combined with end-to-end network platforms to create next generation data centers
Meta Platforms and Elon Musk’s xAI start-up are among companies building clusters of computer servers with as many as 100,000 of Nvidia’s most advanced GPU chips as the race for artificial-intelligence (AI) supremacy accelerates.
- Meta Chief Executive Mark Zuckerberg said last month that his company was already training its most advanced AI models with a conglomeration of chips he called “bigger than anything I’ve seen reported for what others are doing.”
- xAI built a supercomputer called Colossus—with 100,000 of Nvidia’s Hopper GPU/AI chips—in Memphis, TN in a matter of months.
- OpenAI and Microsoft have been working to build up significant new computing facilities for AI. Google is building massive data centers to house chips that drive its AI strategy.
xAI built a supercomputer in Memphis that it calls Colossus, with 100,000 Nvidia AI chips. Photo: Karen Pulfer Focht/Reuters
A year ago, clusters of tens of thousands of GPU chips were seen as very large. OpenAI used around 10,000 of Nvidia’s chips to train the version of ChatGPT it launched in late 2022, UBS analysts estimate. Installing many GPUs in one location, linked together by superfast networking equipment and cables, has so far produced larger AI models at faster rates. But there are questions about whether ever-bigger super clusters will continue to translate into smarter chatbots and more convincing image-generation tools.
Nvidia Chief Executive Jensen Huang said that while the biggest clusters for training for giant AI models now top out at around 100,000 of Nvidia’s current chips, “the next generation starts at around 100,000 Blackwells. And so that gives you a sense of where the industry is moving. Do we think that we need millions of GPUs? No doubt. That is a certainty now. And the question is how do we architect it from a data center perspective,” Huang added.
“There is no evidence that this will scale to a million chips and a $100 billion system, but there is the observation that they have scaled extremely well all the way from just dozens of chips to 100,000,” said Dylan Patel, the chief analyst at SemiAnalysis, a market research firm.
Giant super clusters are already getting built. Musk posted last month on his social-media platform X that his 100,000-chip Colossus super cluster was “soon to become” a 200,000-chip cluster in a single building. He also posted in June that the next step would probably be a 300,000-chip cluster of Nvidia’s newest GPU chips next summer. The rise of super clusters comes as their operators prepare for Nvidia’s nexgen Blackwell chips, which are set to start shipping out in the next couple of months. Blackwell chips are estimated to cost around $30,000 each, meaning a cluster of 100,000 would cost $3 billion, not counting the price of the power-generation infrastructure and IT equipment around the chips.
Those dollar figures make building up super clusters with ever more chips something of a gamble, industry insiders say, given that it isn’t clear that they will improve AI models to a degree that justifies their cost. Indeed, new engineering challenges also often arise with larger clusters:
- Meta researchers said in a July paper that a cluster of more than 16,000 of Nvidia’s GPUs suffered from unexpected failures of chips and other components routinely as the company trained an advanced version of its Llama model over 54 days.
- Keeping Nvidia’s chips cool is a major challenge as clusters of power-hungry chips become packed more closely together, industry executives say, part of the reason there is a shift toward liquid cooling where refrigerant is piped directly to chips to keep them from overheating.
- The sheer size of the super clusters requires a stepped-up level of management of those chips when they fail. Mark Adams, chief executive of Penguin Solutions, a company that helps set up and operate computing infrastructure, said elevated complexity in running large clusters of chips inevitably throws up problems.
The continuation of the AI boom for Nvidia largely depends on how the largest clusters of GPU chips deliver a return on investment for its customers. The trend also fosters demand for Nvidia’s networking equipment, which is fast becoming a significant business. Nvidia’s networking equipment revenue in 2024 was $3.13 billion, which was a 51.8% increase from the previous year. Mostly from its Mellanox acquisition, Nvidia offers these networking platforms:
- Accelerated Ethernet Switching for AI and the Cloud
- Quantum InfiniBand for AI and Scientific Computing
- Bluefield® Network Accelerators
………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………..
Nvidia forecasts total fiscal fourth-quarter sales of about $37.5bn, up 70%. That was above average analyst projections of $37.1bn, compiled by Bloomberg, but below some projections that were as high as $41bn. “Demand for Hopper and anticipation for Blackwell – in full production – are incredible as foundation model makers scale pretraining, post-training and inference, Huang said. “Both Hopper and Blackwell systems have certain supply constraints, and the demand for Blackwell is expected to exceed supply for several quarters in fiscal 2026,” CFO Colette Kress said.
References:
https://www.wsj.com/tech/ai/nvidia-chips-ai-race-96d21d09?mod=tech_lead_pos5
https://www.nvidia.com/en-us/networking/
https://nvidianews.nvidia.com/news/nvidia-announces-financial-results-for-third-quarter-fiscal-2025