BICSI Brief Volume 46, Number 1 | January/February/March 2025
ICT TODAY THE OFFICIAL TRADE JOURNAL OF BICSI
Volume 46, Number 1 January/February/March 2025
ENERGIZING THE AI FRONTIER
PLUS: + A I's Impact on the Energy Grid + O ptimizing High-Throughput Wi-Fi
Freeform Ribbon™ One technology. Multiple options. Tailored to your needs.
Sumitomo Electric Lightwave’s Freeform Ribbon™ allows for dense fiber packing and a small cable diameter with a non-preferential bend axis thereby increasing density in space-constrained applications. Sumitomo Electric’s patented pliable Freeform Ribbon™ construction is designed to both pack densely in small form factor cables while still being capable to transform quickly, by hand ONLY, to splice-ready form similar to standard ribbon for fast and easy 12ct ribbon splicing (for both in-line and fusion splice-on connector splicing applications).
SumitomoElectricLightwave.com
contents
JANUARY/FEBRUARY/MARCH 2025 Volume 46, Issue 1
FROM THE BOARD PRESIDENT 05 Innovating at the Crossroads of Technology, Energy, and Values By David M. Richards, RCDD, NTS, OSP, TECH, CT COVER ARTICLE 06 Energizing the AI Frontier The worlds of energy and data are converging like never before. This convergence, driven by the explosive growth of artificial intelligence (AI) and machine learning (ML) technologies, is pushing existing infrastructure beyond conventional boundaries, and reshaping the very foundations of the digital world. By Seán Adam and Manja Thessin 14 Optimizing High-Throughput Wi-Fi: A Practical Guide to Testing and Troubleshooting Wi-Fi has become the cornerstone of connectivity, connecting everything from smartphones and laptops to smart homes and industrial IoT devices. As the demand for faster, more reliable wireless networks continues to grow, network professionals must have a deep understanding of testing and troubleshooting these complex systems. By Mark Mullins 24 Examining the Impact of AI on the Energy Grid In 2024, the discourse surrounding data centers and their operators underwent a significant transformation, propelled by the burgeoning influence of artificial intelligence (AI). This shift was primarily driven by an exponential surge in energy demand from AI-enhanced data centers. Currently, AI applications are estimated to consume between 10% and 20% of the electricity utilized by data centers, and as adoption across industries grows, this percentage will increase. By Dr. Rebecca Bosco
32 Limited-Energy Hybrid Fiber/Power Cables: Delivering Bandwidth and Power to the Edge The rapid evolution of IoT, sensors, AI, and machine learning is accelerating a seismic shift in building management and industrial automation systems, turning data into insights far beyond traditional expectations. This transformation into smart systems equipped with sensors marks a new era where IT plays a vital role in managing and maximizing their potential across diverse applications. By Gayla Arrindell and Steve Eaves 36 ICT Bonding Infrastructure Design Best Practices In the modern hyper-connected world, the seamless integration of information and communications technology (ICT) with electrical power systems is more crucial than ever. As data centers and enterprise networks grow in complexity, ICT professionals face the challenge of ensuring connectivity and optimal safety and performance. By Bob Faber 44 Values vs. Value: The Race to the Bottom in Telecommunications Infrastructure The telecommunications industry is the cornerstone of global connectivity in the rapidly evolving landscape of technology. But there are ethical implications of the race to the bottom in how materials essential to this infrastructure are sourced. It is time to confront the harsh realities and moral questions that underpin these
production processes. By Elaine Kasperek
SUBMISSION POLICY ICT TODAY is published quarterly by BICSI, Inc. and is sent in digital format to BICSI members and credential holders. ICT TODAY welcomes and encourages submissions and suggestions from its readers. Articles of a technical, vendor-neutral nature are gladly accepted for publication with approval from the Editorial Review Board. However, BICSI, Inc., reserves the right to edit and alter such material for space or other considerations and to publish or otherwise use such material. The articles, opinions, and ideas expressed herein are the sole responsibility of the contributing authors and do not necessarily reflect the opinion of BICSI, its members, or its staff. BICSI is not liable in any way, manner, or form for the articles’ opinions and ideas. Readers are urged to exercise professional caution in undertaking any of the recommendations or suggestions made by authors. No part of this publication may be reproduced in any form or by any means, electronic or mechanical, without permission from BICSI, Inc. ADVERTISING: Advertising rates and information are provided upon request. Contact BICSI for information at +1 813.769.1842 or cnalls@bicsi.org. Publication of advertising should not be deemed as endorsement by BICSI, Inc. BICSI reserves the right in its sole and absolute discretion to reject any advertisement at any time by any party. © Copyright BICSI, 2025. All rights reserved. BICSI and all other registered trademarks within are property of BICSI, Inc.
I
January/February/March 2025
3
THE OFFICIAL TRADE JOURNAL OF BICSI
ADVERTISER’S INDEX McGard. ................................................... 47 mcgard.com/durashield/ Adrian Steel..............................................47 newera.adriansteel.com AFL.............................................Back Cover learn.aflglobal.com Sumitomo.................................Inside Cover SumitomoElectricLightwave.com BICSI INFORMATION DD215........................................................ 31 ANSI/BICSI 003-2024.................................45 APPLIED FIBER SPLICING...........................49
BICSI BOARD OF DIRECTORS Board President David M. Richards, RCDD, NTS, OSP, TECH, CT Board President-Elect William "Bill" Foy, RCDD, DCDC, ESS, NTS, OSP, WD Board Secretary Luke Clawson, RCDD, RTPM, GROL, MBA Board Treasurer Peter P. Charland III, RCDD, RTPM, DCDC, SMIEEE, CET, NTS, ESS, WD Board Director Ninad Desai, RCDD, NTS, OSP, TECH, CT Board Director William “Joe” Fallon, RCDD, ESS Board Director Daniel Hunter, RCDD Board Director Trevor Kleinert, RCDD, NTS, DCDC, TECH, CT Board Director Gilbert Romo Board Director Mark Tarrance, RCDD, RTPM Board Director Jay Thompson, RCDD, NTS Board Director James "Jim" Walters, RCDD, DCDC, OSP, RTPM, PMP, CISSP, GICSP Chief Executive Officer John H. Daniels, CNM, FACHE, FHIMSS, CPHIMS
EDITORIAL REVIEW BOARD Beatriz Bezos, RCDD, DCDC, ESS, NTS, OSP, PE, PMP Jonathan L. Jew F. Patrick Mahoney, RCDD, CDT PUBLISHER BICSI, Inc., 8610 Hidden River Pkwy., Tampa, FL 33637-1000 Phone: +1 813.979.1991 Web: bicsi.org EDITOR Dan Brown, icttodayeditor@bicsi.org
ICT TODAY NEEDS WRITERS ICT Today is BICSI’s premier publication for authoritative, vendor-neutral coverage and insight on next generation and emerging technologies, standards, trends, and applications in the global ICT community. Consider sharing your industry knowledge and expertise by becoming a contributing writer to this informative publication. Contact icttodayeditor@bicsi.org if you are interested in submitting an article.
ADVERTISING SALES +1 813.979.1991 or cnalls@bicsi.org
PUBLICATION STAFF Jeff Giarrizzo, Senior Technical Editor Allen Dean, Senior Technical Editor Mark "Line" Cansino, Project Manager
I
4
ICT TODAY
From BICSI’s Board President David M. Richards, RCDD, NTS, OSP, TECH, CT
INNOVATING AT THE CROSSROADS OF TECHNOLOGY, ENERGY, AND VALUES
Dear Members and Readers, It is with great enthusiasm that I welcome you to the latest issue of ICT Today. As always, we strive to bring you content that not only reflects the cutting-edge developments in our industry but also challenges us to think critically about the role we play in shaping the future of information and communications technology. This issue comes at a pivotal time for our profession. The articles within these pages underscore the dynamism and innovation that continue to define the ICT landscape. They also highlight the growing intersections of technology with energy, infrastructure, and values—areas where our expertise as ICT professionals has never been more critical. Our cover feature, "Energizing the AI Frontier" by Seán Adam and Manja Thessin, and Dr. Rebecca Bosco’s article, "Examining the Impact of AI on the Energy Grid" explore how the surging energy demands of AI are reshaping energy systems, emphasizing the critical role ICT professionals play in driving innovation, sustainability, and efficiency at the intersection of technology and infrastructure. "Limited-Energy Hybrid Fiber/Power Cables: Delivering Bandwidth and Power to the Edge," co-authored by Gayla Arrindell and Steve Eaves, examines how integrating power and data delivery with advanced hybrid cabling and fault-managed power systems is addressing the growing demands of smart technologies and connectivity at the edge. In "ICT Bonding Infrastructure Design Best Practices," Bob Faber examines the critical role of bonding and grounding in modern ICT systems, offering practical guidance rooted in industry standards to enhance safety, performance, and adaptability in a rapidly evolving technological landscape.
Elaine Kasperek’s thought-provoking article, "Values vs. Value: The Race to the Bottom in Telecommunications Infrastructure," challenges us to confront the ethical and quality compromises in sourcing practices, urging the industry to balance cost efficiency with human dignity, environmental responsibility, and long- term sustainability. Finally, Mark Mullins’s "Optimizing High-Throughput Wi-Fi: A Practical Guide to Testing and Troubleshooting" offers essential strategies for ensuring robust and reliable wireless networks by addressing infrastructure integrity, Power over Ethernet support, and interference in complex environments. As ICT professionals, we are uniquely positioned to address the challenges and opportunities that arise from the convergence of technology, energy, and infrastructure. The insights shared by our contributors in this issue not only educate but also inspire us to strive for excellence and innovation in our respective fields. I encourage you to engage deeply with the articles and to share your thoughts and experiences with our community. Let this journal be a catalyst for conversation, collaboration, and progress. Thank you for your continued commitment to advancing our profession. Together, we are shaping the future of ICT and creating a more connected and resilient world. Warm regards,
David M. Richards, RCDD, NTS, OSP, TECH, CT President, Board of Directors
I
January/February/March 2025
5
Energizing the AI Frontier By Seán Adam and Manja Thessin
The worlds of energy and data are converging like never before. This convergence, driven by the explosive growth of artificial intelligence (AI) and machine learning (ML) technologies, is pushing existing infrastructure beyond conventional boundaries, and reshaping the very foundations of the digital world. The current magnitude of energy consumption required by AI is unprecedented. Today’s AI models, growing exponentially in size and complexity, demand massive amounts of computing power and, consequently, energy. This insatiable appetite for power is not just a technical challenge—it is a fundamental issue that is already impacting digital infrastructure and energy systems.
6
I
ICT TODAY
ENERGY SECTOR TRANSFORMATION The energy sector is undergoing a profound transformation, driven by the need to meet the escalating power demands of an increasingly digital world, particularly the voracious appetite of AI and machine learning technologies. This transformation encompasses three key areas: grid modernization, integration of renewable energy sources, and smart grid technologies. Grid Modernization Aging electrical infrastructure, much of which was established in the mid-20th century, was not designed to handle the complex power flows and intermittent nature of renewable energy sources, let alone the massive energy demands of AI data centers. To address this challenge, upgrading power grids with advanced sensors, control systems, and energy storage technologies is crucial to ensure reliable, efficient, and sustainable power delivery. These upgrades will enable real-time monitoring, predictive maintenance, and dynamic load balancing— essential features for managing the fluctuating energy needs of AI workloads.
The industry is also investing in advanced grid technologies (Figure 1) that enable utilities to increase the transmission capacity of their existing infrastructure and rights-of-way. These include advanced transmission technologies such as advanced conductor and grid-enhancing technologies (GETs) like dynamic line rating, virtual power plants, and utility- scale long-term storage. Integration of Renewable Energy Sources The shift towards renewable energy sources like solar, wind, and hydroelectric power is a key driver of the energy sector’s transformation. While these clean energy sources offer a path to improved sustainability, they also present challenges due to their location and their variable and intermittent nature. Unlike traditional fossil fuel-based generation, which can be stood up close to the demand, renewable energy resources need to be built where there are sufficient natural resources (e.g., wind, sunlight). This requires utilities to build out new transmission infrastructure, often across hundreds of miles, to deliver power to the point of consumption.
FIGURE 1: Overview of Advanced Technologies and Applications (Source: DOE Liftoff Report). 1
January/February/March 2025
I
7
• Demand response: Data centers can participate in demand response programs, adjusting their power consumption during peak periods to help balance the grid and potentially reduce energy costs. • Improved reliability: Advanced monitoring and predictive maintenance capabilities can help prevent outages and ensure a stable power supply for critical AI operations. Integrating smart grid technologies with AI-driven energy management systems within data centers creates a symbiotic relationship between energy providers and consumers, balancing energy use across the entire ecosystem. INFRASTRUCTURE CHALLENGES The rapid advancement of AI technologies is pushing data center infrastructure into uncharted territory, creating unprecedented challenges in bandwidth, latency, power, and cooling. These challenges are reshaping the foundations of data center design and operation, demanding innovative solutions and a fundamental rethinking of the approach to digital infrastructure. Bandwidth and Latency AI workloads are driving an unprecedented demand for low latency and high bandwidth connectivity between servers, storage, and GPUs. The impact on network performance is substantial. According to Meta, on average, 33% of AI elapsed time is spent waiting for the network. 4 Such delays incur timeouts that affect customer service, increase costs, and impede scalability. To address these challenges, data centers are exploring advanced networking solutions. The industry is rapidly moving towards higher speed interfaces, with 800 Gbps expected to make up most AI back-end network ports by 2025, according to Dell’Oro Group (Figure 2). Some companies are even pushing beyond, with Synopsys debuting a 1.6T Ethernet IP core designed to meet the high bandwidth needs of AI and hyperscale data center chips. 5
It is estimated that over 2.6 terawatts of energy generation capacity are in the transmission interconnect queue, with a 4-5 year wait time for approval and build. 2 For AI data centers, which require consistent and substantial power, the integration of renewables poses additional challenges. While renewable energy can significantly reduce the carbon footprint of these energy-intensive facilities, the variability requires sophisticated energy storage and management systems to ensure uninterrupted power supply. To address this challenge, the industry is investing heavily in advanced clean energy solutions such as next-generation geothermal, advanced nuclear, clean hydrogen, and long-duration energy storage. 3
Smart Grid Technologies The third pillar of this transformation is the development and deployment of smart grid technologies. These advanced systems leverage AI and IoT devices to create a more intelligent, responsive, and efficient power grid. Smart grids enable bidirectional communication between utilities and consumers, allowing for real-time adjustments to power distribution based on demand and availability. Smart grid technologies offer several benefits for AI data centers: • Dynamic load balancing : Smart grids can automatically adjust power distribution to meet the fluctuating demands of AI workloads, ensuring optimal performance while minimizing energy waste.
I
8
ICT TODAY
FIGURE 2: Migration to High-Speeds in AI Clusters (AI Back-End Networks) (Source: Dell'Oro Group AI Networks Report Dec 2023) 6
Power and Cooling Modern AI hardware has server rack power demands ranging from traditional 4-6kW up to an astonishing 50-100kW or more, with next-generation configurations expected to require between 60kW to 120kW per rack. 7 Managing heat dissipation in these high-density racks is a critical challenge. Traditional air-cooling methods are insufficient to handle the heat generated by AI-focused hardware. To address these cooling challenges, the industry is turning to innovative solutions like liquid cooling. Liquid cooling technologies offer several advantages, including improved energy efficiency, reduced water consumption, and the ability to handle higher power densities. Some liquid cooling solutions can remove nearly 100% of the heat generated across the entire IT stack while reducing energy use by up to 40% and water consumption by up to 100%. 8 DATA INFRASTRUCTURE EVOLUTION The widescale adoption of AI technologies is driving a fundamental transformation in data infrastructure. This transformation includes how data is collected,
stored, processed, and analyzed—creating a new paradigm in data center design and operation.
Network Architecture Changes The exponential growth in AI model complexity is pushing data infrastructure to its limits. AI models are growing by a factor of 1,000 every three years, requiring extreme scale and bandwidth for efficient operation. 9 This growth has fundamentally changed data center traffic patterns, driving a shift from traditional north-south traffic patterns (data moving between servers and the internet) to east-west traffic (extensive communication between servers and nodes) within data centers. This architectural shift has significant energy implications. The increased east-west traffic requires more internal networking infrastructure and creates additional heat loads that must be managed. Network delays and idle computing resources further contribute to increased energy consumption, making efficient network architecture imperative for both performance and energy optimization.
January/February/March 2025
I
9
Next-generation Data Centers The debate between distributed vs. centralized architectures is intensifying, driven by competing demands of scale and latency. While centralized data centers offer significant economies of scale for computing and power delivery, AI applications' latency requirements push workloads toward edge computing and distributed architectures. The evolution toward distributed architectures presents unique energy challenges. Edge nodes must balance the benefits of processing data closer to its source—which can reduce transmission energy costs—against the potential loss of energy efficiency from operating at a smaller scale. This has led to the development of hybrid models that look to optimize both energy use and performance. These hybrid models represent a new approach to data center design. They implement advanced networking solutions with dynamic resource allocation that can adjust power consumption based on traffic patterns and workload distribution. For example, edge nodes handle time-sensitive AI inference tasks, while more complex, energy-intensive training workloads are offloaded to centralized facilities where power delivery and cooling infrastructure can be optimized at scale. This adaptive approach ensures more efficient computing and energy resources are utilized across the distributed architecture. FUTURE OUTLOOK At the intersection of data and energy, the future of digital infrastructure is being shaped by three key trends: more sustainable data infrastructure, the development of integrated energy-data systems, and technological convergence across multiple domains. These trends are not just shaping the future of ICT—they are redefining the foundations of the digital economy and society. Sustainable Data Infrastructure The imperative for improving sustainability is driving innovation in data center design and operation. Green energy solutions are no longer just a nice-to-have; they are becoming a customer and government requirement for data centers, especially those powering AI workloads.
This shift is not just about environmental responsibility— it is about long-term viability in a world of increasing energy costs and regulatory pressures. Solar and wind technologies are improving in efficiency and energy storage solutions are helping to mitigate their intermittent nature. However, the massive energy demands of AI data centers mean that these renewable sources alone may not yet be sufficient. This reality is pushing the industry to explore a diverse range of energy solutions: Geothermal energy offers a consistent power source but faces challenges in the cost of implementation and location limitations.
Bioenergy presents another alternative, though it requires careful management of biomass sources and processing.
Nuclear energy is re-emerging as a potential solution for powering large AI data centers (Figure 3). Small Modular Reactors (SMRs), producing around 300 MW(e) per unit, could provide the low-carbon, high- density energy needed for these facilities. While public perception and safety concerns remain significant hurdles, the reliability and energy density of nuclear power make it a compelling option for meeting the consistent, substantial power needs of AI infrastructure. The future of sustainable data infrastructure will likely involve a mix of these technologies, tailored to local conditions and regulatory environments. This diversification will not only enhance sustainability but also improve resilience and energy security.
Integrated Energy-Data Systems The future will see an increasingly symbiotic
relationship between power grids and data networks. Building on current smart grid technologies, future developments will usher in “energy-aware computing,” where data centers will not just be consumers in the energy ecosystem.
I
10
ICT TODAY
Technological Convergence The convergence of advanced networking solutions, energy-aware data management, and next-generation data center architectures is set to reshape the ICT landscape. This convergence is not just about combining existing technologies—it is about creating new capabilities. In networking, there is a persistent push towards higher speeds and lower latency: • Optical interconnects for AI clusters are becoming increasingly common, enabling the massive data transfers required for distributed AI training. • Coherent optics extend high-speed connections over long distances, facilitating data center interconnects and edge computing deployments. • Emerging fiber technologies like Multi-Core Fiber (MCF) and Hollow-Core Fiber (HCF) promise to increase bandwidth while reducing latency. These technologies could be game changers for AI applications that require real-time processing of vast amounts of data.
Key future developments in this area may include:
• Advanced dynamic power management Data centers could dynamically adjust their power consumption based on grid conditions and energy pricing, potentially feeding excess power from periods of lower demand back into the grid. • Next-generation AI-driven energy optimization Future AI systems may constantly analyze and optimize energy use across the entire data center ecosystem, including adjusting cooling systems based on workload predictions and orchestrating workloads across different geographic locations.
• Energy-data marketplaces The emergence of sophisticated energy-data
marketplaces, where data centers can trade compute capacity and energy resources in real-time, would optimize performance and sustainability.
FIGURE 3: Nuclear reactors can range in size and output. (Image: A. Vargas/IAEA) 10
January/February/March 2025
I
11
Energy-aware data management is growing more sophisticated with real-time monitoring and control systems, powered by AI and machine learning, enabling greater levels of energy optimization. These systems can predict workload patterns, adjust cooling systems in real-time, and even make decisions about where to process data based on energy availability and cost. This technological convergence is not just about incremental improvements, it is about creating a new, intelligent infrastructure that can self-govern and adapt to changing demands and conditions. As these technologies mature, the emergence of a truly smart, energy-efficient, and high-performance digital ecosystem will follow. INDUSTRY IMPACT The convergence of energy and data is driving significant changes in standards and regulations and creating new market opportunities. How the industry operates and plans for the future will create both challenges and opportunities for ICT professionals. Standards and Regulations As data centers become increasingly energy-intensive, particularly with the rise of AI workloads, regulatory bodies are responding with demands for new standards and regulations. These evolving standards are helping the industry move towards more sustainable practices. For instance, the European Union has mandated that data centers in member countries must be climate-neutral by 2030. 11 This ambitious target is driving innovation in energy-efficient technologies and renewable energy integration. In the United States, the Department of Energy is developing new efficiency standards for data center equipment, which will likely impact the industry's approach to hardware design and deployment. The ICT industry must stay ahead of regulatory changes, not just to ensure compliance but to protect future innovation. Proactive companies that adopt sustainable practices and energy-efficient technologies will be better positioned to meet future regulations and gain a competitive edge.
Interoperability in AI-driven optical networks is another area where standards are rapidly evolving. As data centers adopt advanced networking technologies to meet the bandwidth demands of AI workloads, ensuring seamless integration and communication between different systems becomes crucial. Industry bodies like the IEEE and ITU are working on standards for coherent optics and software-defined networking, which will be essential for the next generation of data center networks. Market Opportunities The convergence of energy and data is opening significant market opportunities, particularly in the transmission and substation sectors. The global transmission substation market size crossed $81.7 billion in 2023 and is projected to grow at a CAGR of 2.6% from 2024 to 2032. 12 This growth is driven by the increasing global demand for electricity and the rise of renewable energy sources. Emerging technologies are driving infrastructure upgrades across the industry, with the push for smart grids and digital substations creating new opportunities for companies specializing in sensors, control systems, and data analytics. The U.S. Department of Energy's Grid Deployment Office, with about $26 billion in available funding, is accelerating investments in grid modernization and resilience. 13 Additionally, the rise of edge computing and distributed energy resources is creating new markets. This convergence at the edge presents unique opportunities for companies that can provide integrated solutions for managing both energy and data. Edge computing solutions that optimize for both performance and energy efficiency will be in high demand, and there is potential for new business models that leverage the synergies between distributed energy resources and edge computing infrastructure. For ICT professionals, these trends present exciting opportunities for skill development and career growth. Expertise in areas such as energy-efficient data center design, smart grid technologies, and edge computing will be invaluable in the coming years.
12
I
ICT TODAY
CONCLUSION The convergence of energy and data represents an inflection point for the ICT industry. With the exascale era on the horizon, and AI models and data centers projected to grow to unprecedented sizes, daunting challenges and exciting opportunities will emerge. This convergence is not just a technical challenge— it is a fundamental shift in digital infrastructure that will have far-reaching implications for society, the economy, and the environment. The future energy demands of AI will impact power generation, distribution, and consumption. A transformation of energy infrastructure, from advanced grid modernization to the integration of diverse renewable sources, is anticipated. Simul- taneously, data centers will need to meet the extreme bandwidth, latency, and processing requirements of future AI workloads. This convergence is expected to drive innovation across multiple fronts, from sustainable data infra- structure to integrated energy-data systems. The ICT industry stands at the forefront of this industrial revolution, with the unique opportunity—and responsibility—to lead the way in creating a more energy-efficient, sustainable digital future. AUTHOR BIOGRAPHIES: Seán Adam is Vice President of Market Strategy and Innovation at AFL. Seán holds a master's degree in Electric Engineering from the Massachusetts Institute of Technology and has over 25 years of experience in the Semiconductor and Telecom industry with a focus on system- based solutions and architectures. Seán holds over 15 patents across software and hardware systems and products. Today, Seán is focused on developing and guiding AFL's overall growth strategy in the key markets they serve. Learn more about Seán on LinkedIn (www.linkedin.com/in/seanadam) or follow him on Twitter (@AFL_SeanAdam). Manja Thessin , RCDD/RTPM serves as Market Manager for AFL. Her nearly 25 years of experience in the telecommunications industry spans fieldwork, design and engineering, and project management. Manja’s expertise is further enhanced by a master's certificate in Strategic
Leadership from Michigan State University, as well as RCDD and RTPM certifications from BICSI ® . Her current role focuses on strategic initiatives that address global challenges and unlock potential for the organization, leveraging her deep understanding of ICT and her talent for cross-organizational collaboration. References: 1. U.S. Department of Energy. (2024). Pathways to Commercial Liftoff: Innovative Grid Deployment. Washington, DC: DOE. 2. Wolfe, S. (2024, April 10). U.S. interconnection queues, already jammed, grew 30% in 2023. Renewable Energy World. 3. Texier, M. (2024, March 19). A new initiative with Microsoft and Nucor to accelerate advanced clean electricity technologies. Google Blog. 4. Boujelbene, S. (2023, October 30). OCP 2023 Key Takeaways: The Network is the Computer for AI Workloads. Dell'Oro Group. 5. Synopsys. (2024, February 29). Synopsys Launches Industry's First Complete 1.6T Ethernet IP Solution to Meet High Bandwidth Needs of AI and Hyperscale Data Center Chips. 6. Boujelbene, S. (2024, January 10). Exploring the Data Center Switch and AI Networks Markets Landscape in 2024. Dell'Oro Group. 7. Lear, D. (2024, April 19). Top 5 Considerations for Deploying NVIDIA Blackwell. AMAX. 8. Adams, H. S. (2024, August 2). Iceotope's Liquid Cooling & Edge Computing Efficiency. DataCentre Magazine. 9. Volico. (2024, April 2). How Data Centers Are Adapting To The Rise Of AI. 10. International Atomic Energy Agency. (2023, September 13). What are Small Nuclear Reactors (SMRs)? IAEA. 11. Climate Neutral Data Center Pact. (2023). Retrieved from CISPE.cloud. 12. Global Market Insights. (2024, July). Transmission Substation Market Size, 2024 - 2032. 13. U.S. Department of Energy Grid Deployment Office. (2023, November 22). Where are we now? Celebrating the Bipartisan Infrastructure Law's Historic Investment in Critical Electric Infrastructure.
I
I
13
ICT TODAY
January/February/March 2025
13
Optimizing High-Throughput Wi-Fi: A Practical Guide to Testing and Troubleshooting
By Mark Mullins
Wi-Fi has become the cornerstone of connectivity, connecting everything from smartphones and laptops to smart homes and industrial IoT devices. As the demand for faster, more reliable wireless networks continues to grow, network professionals must have a deep understanding of testing and troubleshooting these complex systems. Based on IEEE 802.11 standards, Wi-Fi has evolved rapidly over the years to offer increasingly higher speeds via new operating frequencies, more spatial streams, and advanced encoding and beamforming technologies. Table 1 compares Wi-Fi 5, Wi-Fi 6, Wi-Fi 6E, and the latest Wi-Fi 7 generations.
TABLE 1: Comparison of Wi-Fi generations (Source: Fluke Corporation)
I
14
ICT TODAY
Every Wi-Fi access point (AP) connects to a LAN with Ethernet cabling, typically balanced twisted-pair copper cabling. Most Wi-Fi access points also receive power over that cabling via Power over Ethernet (PoE) technology. Depending on the generation, wireless APs use radio frequency (RF) signals operating in the 2.4 GHz, 5 GHz, and/or 6 GHz frequency bands to communicate with devices. Network installers and technicians therefore have multiple considerations when testing and troubleshooting Wi-Fi to optimize performance and provide a seamless user experience. CERTIFYING INFRASTRUCTURE INTEGRITY The performance of the wireless network will be no better than the performance of the cabling connecting it. Cabling issues can cause intermittent signal loss and poor performance which are difficult to diagnose. Certification testing is crucial for ensuring the reliability and performance of structured cabling installations by demonstrating compliance with applicable industry cabling standards. Most cabling system manufacturers even require certification testing to acquire a system warranty. Certification test results also protect installers from post-installation issues that
are no fault of their own and provide a valuable benchmark for troubleshooting. Certification testing for twisted-pair cabling assesses specific performance parameters, including:
• Insertion loss: Measures signal attenuation over the cable length.
• Near-end crosstalk (NEXT) and power sum NEXT (PSNEXT): Quantifies interference between pairs within a cable. • Near-end attenuation to crosstalk ratio (ACR-N) and power sum ACR-N (PSACR-N): Evaluates the balance between signal strength and crosstalk. • Far-end attenuation to crosstalk ratio (ACR-F) and power sum ACR-F (PSACR-F): Assesses the balance between signal strength and crosstalk at the far end of the cable.
• Return loss: Measures the amount of signal reflected to the transmitter.
I
January/February/March 2025
15
high, it can reduce the PoE power available for a device. DC resistance unbalance evaluates the difference in resistance between conductors and pairs. Too much DC resistance unbalance can cause Ethernet signals to become distorted, resulting in bit errors, retransmits, and even nonfunctioning data links. Measuring DC resistance unbalance between pairs is especially critical in IEEE 802.3bt four-pair PoE (Type 3 and Type 4), which transmits power and data over all four pairs of a twisted-pair cable. IEEE 802.3 PoE standards specify a maximum DC resistance unbalance of 3 percent between conductors, meaning that the difference in DC resistance between two conductors is no more than 3 percent of the total DC loop resistance of the pair. The standards specify a maximum DC resistance unbalance of 7 percent between two pairs, meaning that the difference in parallel resistance is no more than 7 percent of the total parallel resistance of the two pairs. DC resistance unbalance between pairs requires six measurements to compare each pair to every other pair. Figure 1 shows the DC resistance unbalance between conductors for Pair 1,2 and Pair 4,5, and the DC resistance unbalance between both pairs. In this example, DC resistance unbalance passes for Pair 4,5 but fails for Pair 1,2. The cabling also fails pair-to-pair DC resistance unbalance.
Category 6A/Class FA cabling that supports 10 Gb/s Ethernet (i.e., 10GBASE-T) often also requires alien crosstalk testing, including power sum alien near-end crosstalk testing (PSANEXT) and power sum alien attenuation to crosstalk ratio far-end (PSAACR-F). This is due to higher-frequency signals causing interference from adjacent cables. Industry cabling standards recommend two Category 6A/Class FA connections to Wi-Fi 6/6E APs and four Category 6A/Class FA connections to Wi-Fi 7 APs. Depending on the project specification, Wi-Fi APs connected via fiber also necessitate Tier 1 certification testing with an optical loss test set (OLTS) that measures insertion loss on a link, or Tier 2 certification testing using an optical time domain reflectometer (OTDR) that characterizes the loss of the link for individual splices and connectors by transmitting light pulses into a fiber and measuring the amount of light reflected from each pulse. ENSURING SUPPORT FOR PoE While not typically required for certification of installed cabling, extended field testing of direct current (DC) loop resistance and DC resistance unbalance can help ensure adequate support for PoE delivery to wireless APs and other PoE-enabled devices. In twisted pair cabling, DC loop resistance measures the total resistance of a pair. If DC resistance is too
FIGURE 1: DC resistance unbalance for Pair 1,2 and Pair 4,5 and between Pair 1,2 and Pair 4,5 (Source: Fluke Corporation)
I
16
ICT TODAY
While DC loop resistance and DC resistance unbalance can be a result of low-quality cables, DC resistance unbalance can also be caused by poor workmanship, such as inconsistent conductor terminations, exceeding bend radius requirements, or failure to maintain pair twists as close to the point of termination as possible. Extending certification testing to include DC resistance unbalance requires a certification tester capable of testing these parameters in the field. Testers with this capability automate the calculations to provide a PASS or FAIL rating. Ensuring support for PoE via DC resistance unbalance testing will become increasingly vital for newer Wi-Fi 7 APs with more spatial streams that demand more power via four-pair Type 4 PoE. TROUBLESHOOTING WI-FI: A STEP-BY-STEP GUIDE Even if the cabling infrastructure passed certification testing at the time of installation, including DC resistance unbalance testing, problems with Wi-Fi can still arise once the network is up and running. With wide-scale reliance on Wi-Fi across the enterprise and beyond, network operators must quickly identify and address complaints such as slow speeds, inability to connect, and dropped connections.
Problems with Wi-Fi are often blamed on cabling. While damaged cables can certainly cause issues, the problem might lie elsewhere in the network infrastructure. The following steps walk through a systematic approach to troubleshooting Wi-Fi problems, focusing on key areas like cabling, network configuration, PoE, and the Wi-Fi environment itself. STEP 1: VERIFY CABLING INTEGRITY The first step in troubleshooting Wi-Fi is to verify the connectivity of the link. This is achieved through simple wire map testing for opens, shorts, and miswirings. Using a cable tester with graphical display and analysis capabilities can simplify wire map testing (Figure 2). When wire map testing reveals an open and potentially damaged cable, locating the break can be difficult if cables are not easily accessible (i.e., behind walls or in conduit). Testers with a built-in tone generator are ideal for locating a break. For cabling infrastructure comprised of shielded twisted pair cabling, the problem could be a break in the continuity of the shield. Therefore, a cable tester with the ability to test shield integrity is a consideration.
FIGURE 2: Cable testers that graphically display and analyze wire map test results can facilitate identifying opens, shorts, or miswirings that prevent signals from reaching a connected device. (Source: Fluke Corporation)
I
January/February/March 2025
17
that characterizes the loss of the link for individual splices and connectors by transmitting light pulses into a fiber and measuring the amount of light reflected from each pulse. STEP 3: ASSESS PoE OPERATION While adding DC resistance unbalance testing to the certification testing process can ensure support of two- and four-pair PoE applications for powering Wi-Fi APs, once the network goes live, there are various reasons why a PoE-enabled Wi-Fi AP may not receive sufficient power. On a PoE-enabled switch, power is typically allocated per port based on the device requirements. For example, a 24-port PoE-enabled switch may have a power budget of 740 W, enabling 30 W of PoE on each port. With such a wide variety of PoE devices residing on the network, it is unlikely that every connected device will need the full 30 W—a surveillance camera on one port might only need 12 W, while a high-throughput Wi-Fi 6 AP on another might need 25 W. Power allocation can be used to adjust port power and improve overall efficiency. The basic rule is that for a connected device to receive adequate power, the PoE class delivered via the switch port must be equal to or greater than the class of its connected powered device (PD). The total power required for all the devices cannot exceed the total power budget of the switch. Table 2 shows the various PoE classes for Type 1, Type 2, Type 3, and Type 4 PoE.
Confirming that the cable can support the application (e.g., 1000BASE-T) that the Wi-Fi AP requires is essential. While some simple testers will verify continuity, a true qualification tester uses specific standards-based measurements to test the cable across a range of frequencies to determine application support. Qualification testing is a quick and easy way to determine if existing Category 5e/Class D or Category 6/Class F can support 2.5 Gb/s or 5 Gb/s for Wi-Fi 5 devices (see sidebar). STEP 2: TEST THE NETWORK Once any cabling issues have been ruled out, the next step is to ensure the Wi-Fi AP is connected to the correct switch port and that the switch is configured properly. For example, links assigned to the wrong virtual LAN may not be able to communicate. Ethernet network switches use link layer discovery protocol (LLDP) or Cisco discovery protocol (CDP) to discover connected devices and advertise their capabilities. Network testers with the ability to leverage these protocols can display the switch name, port, advertised speeds, and assigned VLAN. Some testers even include a blink port light feature to help locate the connected switch port. An Internet Protocol (IP) ping test is also ideal for determining the accessibility of devices such as the wireless controller. Ping tests can also check for latency by determining the maximum round trip time through the network. Category 6A/Class FA cabling that supports 10 Gb/s Ethernet (i.e., 10GBASE-T) often also requires alien crosstalk testing, including power sum alien near-end crosstalk testing (PSANEXT) and power sum alien attenuation to crosstalk ratio far-end (PSAACR-F). This is due to higher-frequency signals causing interference from adjacent cables. Industry cabling standards recommend two Category 6A/Class FA connections to Wi-Fi 6/6E APs and four Category 6A/Class FA connections to Wi-Fi 7 APs. Depending on the project specification, Wi-Fi APs connected via fiber also necessitate Tier 1 certification testing with an optical loss test set (OLTS) that measures insertion loss on a link, or Tier 2 certification testing using an optical time domain reflectometer (OTDR)
“The basic rule is that for a connected device to receive adequate power, the PoE class delivered via the switch port must be equal to or greater than the class of its connected powered device (PD).”
I
18
ICT TODAY
TABLE 2: PoE types, standards, and classes with maximum power delivered by the power sourcing equipment (PSE) and maximum allowed power at the powered device (PD). (Source: Fluke Corporation)
Ethernet Alliance-approved devices provide a simple way to ensure adequate power is supplied. Power Sourcing Devices (PSEs) and Powered Devices each feature a simple label indicating the amount of power they can supply or require. Checking that the switch's power class is greater than the AP's ensures adequate power will be available. This certification also ensures that the devices comply with Ethernet standards. However, while today's advanced PoE switches can determine power requirements of connected devices via power-negotiation messages and allocate power appropriately, problems can arise if that communication malfunctions, PoE has been manually (and incorrectly) allocated per port, or the overall power budget of the switch is oversubscribed. Testers leveraging LLDP and CDP for network testing can also use these protocols to determine the advertised PoE power on a switch port. Testers that perform PoE load testing go a step further by indicating the specific negotiated power class (i.e., 0 through 8) and actual power provided by the switch port. Because power negotiation between a switch and a device happens both at the hardware and software levels, a PoE load tester should ideally display both hardware and software negotiated power information. Negotiated power for a device greater than what a switch port can deliver, whether due to improper allocation, or an oversubscribed power budget, will prevent a PoE- enabled Wi-Fi AP from functioning properly.
To fully understand what's happening, a PoE load tester can indicate which cable pairs carry power. While a single-signature device maintains the same power signature across Pairs 1,2 and 3,6 and Pairs 4,5 and 7,8, a dual-signature device can have independent power signatures on each pair set. Knowing which pairs carry power is helpful, because IEEE 802.3bt PoE (Type 3 and Type 4) supports both topologies. If a switch only supports single-signature devices, it may need to be updated to support dual-signature devices. Such testers can also measure the amount of power under load to confirm that the sourced power is reaching the AP and not lost due to poor-quality cable or a cable link that is too long. STEP 4: CHECK AP CONFIGURATION Recent advancements in Wi-Fi technology have introduced more operating frequencies, channels, channel widths, and security protocols. This increased complexity can lead to configuration issues unrelated to cabling, networking, or PoE operation. The Federal Communications Commission (FCC) and the International Telecommunication Union (ITU) allocate the 2.4 GHz, 5 GHz, and 6 GHz frequency bands for Wi-Fi. Within these bands, specific 20 MHz-wide channels are designated for Wi-Fi. As shown in Figure 3, the 2.4 GHz band has 14 channels, the 5 GHz band has 29, and the 6 GHz band has 59.
I
January/February/March 2025
19
FIGURE 3: Channel numbers designated for Wi-Fi use within the 2.4 GHz, 5 GHz, and 6 GHz frequency bands for the United States. (Source: Fluke Corporation)
While the 5 GHz and 6 GHz bands offer non- overlapping channels, most 2.4 GHz channels overlap, increasing the potential for interference and reduced transmission speeds. Only channels 1, 6, and 11 in the 2.4 GHz band are non-overlapping, so the others are not recommended. The non-overlapping nature of 5 GHz and 6 GHz channels enables bonding multiple smaller 20 MHz channels into wider 40 MHz, 80 MHz, or 160 MHz channels to improve throughput. Some Wi-Fi channels are restricted or designated for specific uses, which can vary by region. For example, channels 12 and 13 of the 2.4 GHz band in North America can only operate in low power mode to avoid interfering with satellite phones. Channel 14 in the 2.4 GHz spectrum is prohibited in North America due to military use. In addition, channels 68 through 96 of the 5 GHz band are not allocated for unlicensed use in North America. Understanding channel usage regulations based on your region is crucial for configuring Wi-Fi APs to avoid restricted channels. Outdated protocols, outdated firmware, disabled encryption, or weak passwords can result in overall weak Wi-Fi security. Wireless security protocols have
evolved significantly over the years. WEP, the initial security protocol introduced in 1997, is now obsolete due to known security vulnerabilities. WPA and WPA2 offered improvements, but WPA3 is the latest recommended security protocol. It provides the most robust protection through individualized encryption rather than shared passwords and a secure handshake to enhance authentication. Configuring all APs to use the WPA3 protocol is highly recommended. STEP 5: ANALYZE THE RF ENVIRONMENT When troubleshooting has ruled out the previously mentioned issues, it is time to analyze the RF environment. The three main factors to investigate are interference, signal strength, and utilization. RF-capable Wi-Fi testers that analyze these three factors within each frequency band and indicate AP-specific data like MAC address, encryption protocol, and channel are essential for understanding and troubleshooting the RF environment. Because RF environments are dynamic, these tools are also handy for ongoing Wi-Fi assessments to maintain performance.
I
20
ICT TODAY
Page 1 Page 2 Page 3 Page 4 Page 5 Page 6 Page 7 Page 8 Page 9 Page 10 Page 11 Page 12 Page 13 Page 14 Page 15 Page 16 Page 17 Page 18 Page 19 Page 20 Page 21 Page 22 Page 23 Page 24 Page 25 Page 26 Page 27 Page 28 Page 29 Page 30 Page 31 Page 32 Page 33 Page 34 Page 35 Page 36 Page 37 Page 38 Page 39 Page 40 Page 41 Page 42 Page 43 Page 44 Page 45 Page 46 Page 47 Page 48 Page 49 Page 50Made with FlippingBook - Online catalogs