Data Center Power & Cooling
The energy side of the AI buildout: rack power densities, liquid cooling, on-site generation, grid constraints, and sustainable data center design.
142 video interviews · 2022–2026
MicroLEDs for AI Data Center Connectivity at Scale
- MicroLED links use dense arrays of modest-speed LEDs rather than fast lasers
- Spare-channel failover delivers reliability and resilience
- Power consumption is in the low single-digit picojoules per bit
Full summary
LK Bhupathi, AVP, Product at Credo, discusses the company's expansion into microLED-based technologies. He explains that microLED solutions use dense arrays of LEDs operating at modest speeds to deliver superior reliability, resilience through spare channel failover, and low single-digit pJ per bit power consumption.
Solving Power & Speed Challenges at 150-300kW Per Rack
- 9-12 month technology cycles outpace 18-24 month infrastructure builds — ECL compresses both
- Rack power densities are reaching 150-300kW
- A 'flex grid' approach combines multiple energy sources per site
Full summary
Yuval Bachar, Co-founder and CEO of ECL, discusses how his AI data center company addresses the mismatch between rapid 9-12 month technology cycles and slower 18-24 month infrastructure timelines by building equipment and facilities on accelerated schedules while managing increasingly power-intensive racks reaching 150-300 kilowatts through a "flex grid" approach combining multiple energy sources.
Lightmatter's Optical Interconnects for AI Scale-Up
- Passage EVK50 delivers DWDM with 16 wavelengths per fiber at 400Gbps Tx + 400Gbps Rx
- Architecture aligns with the new OCI MSA (8-wavelength, 4+4 band interleave)
- Targets energy-efficient, compact optical scale-up interconnects
Full summary
Steve Klinger, VP of Product at Lightmatter, presents the company's optical scale-up interconnect solutions, featuring the Passage EVK50 system with DWDM technology that delivers 16 wavelengths per fiber with 400Gbps Tx and 400Gbps Rx. The architecture offers high energy efficiency and compact integration while aligning with the recently announced OCI MSA specifications using an 8-wavelength, 4+4 band interleave model.
The Optics Scaling Challenge at 1.6T
- GPU interconnect speeds double each generation, forcing copper-to-optical transitions inside the rack
- Power and component supply are the binding constraints on AI data center optics
- Nokia positions its indium phosphide (InP) technology as an open component supply for hyperscalers
Full summary
Neel Patel, GM, Optical Networking Component Solutions at Nokia, explains in our Director's Cut edition, how AI data centers face significant power and supply challenges as GPU interconnect speeds double with each generation, requiring transitions from copper to optical solutions within racks. He highlights Nokia's competitive advantages in indium phosphide (InP) technology, positioning the company as an open ecosystem component supplier for hyperscalers.
Linear Pluggable Optics for Data Center Efficiency
- 1.6T optical modules consume nearly double the power of 800G — up to +1kW on a 64-port switch
- Linear Pluggable Optics (LPO) cuts power by leaning on the switch ASIC's DSPs
Full summary
Neel Patel, GM of Optical Networking Component Solutions at Nokia, discusses the power consumption challenges of advancing optical modules in data centers, where 1.6T modules consume nearly double the power of 800G modules, potentially adding 1KW to a 64-port switch. He explains how Linear Pluggable Optics (LPO) addresses this by leveraging switch ASIC DSPs reducing power consumption.
AI-Driven Multi-Die Design
- Multi-die design adoption is accelerating for AI silicon
- Advanced packaging enables larger formats and denser interconnect
- Power delivery, thermal, and multiphysics challenges now exceed manual design methods
Full summary
Abhijeet Chakraborty, VP of Engineering at Synopsys, examines the accelerating adoption of multi-die designs for AI applications, highlighting how advanced packaging enables larger formats and better interconnect density while presenting complex engineering challenges in power delivery, thermal management, and multiphysics modeling that exceed the capabilities of traditional manual methods.
EPB’s Commercial Quantum Network
- EPB launched the first commercially available quantum network in the U.S. in 2023
- Chattanooga municipal utility partnered with the Department of Energy and Qubitekk
- The network lets industry, academia, and government test quantum tech on a live network
Full summary
Evann Freeman, Director of Government Relations at EPB, discusses how the Chattanooga municipal utility launched the first commercially available quantum network in the U.S. in 2023, partnering with the Department of Energy and Qubitekk to enable industry, academia, and government to test quantum technologies on a live network.
The New Open Centralized Unit Distributed Unit (OCUDU) Effort
- Wireless vendors are shifting to software differentiation atop standardized CPUs and GPUs
- Linux Foundation OCUDu and OCP projects open the door to hardware harmonization
- Marvell claims to be the only provider of macro-grade merchant silicon, trading custom silicon for better cost and power at scale
Full summary
Joel Brand, AVP, Product Marketing at Marvell, explains how major wireless vendors are shifting to software differentiation while adopting standardized CPUs and GPUs, creating opportunities for harmonization through initiatives like the Linux Foundation's OCUDu and OCP projects. He highlights that Marvell is uniquely positioned as the only provider of macro-grade merchant silicon to address the industry's economic challenges by moving from custom silicon to market-optimized solutions that offer better cost efficiency and power performance at scale.
Physical AI in Action
- Nokia's physical AI push features Surf delivery trucks with Nvidia Jetson GPUs across seven US cities via DoorDash and Uber Eats
- Nokia, Nvidia, and T-Mobile are building an "AI grid" distributing processing across devices, AI RAN, and network layers
- Enables safe navigation, human assistance, and multilingual support via LLMs running on the radio network
Full summary
Houman Modarres, Head of Enterprise Marketing at Nokia, presents the company's physical AI initiative at Mobile World Congress, showcasing Surf food delivery trucks with Nvidia Jetson GPUs deployed across seven US cities in partnership with DoorDash and Uber Eats. Nokia is collaborating with Nvidia and T-Mobile to build an "AI grid" that distributes processing across devices, AI RAN, and network layers, enabling applications like safe navigation, human assistance, and multilingual customer support through LLMs running on the radio network.
Nokia's 800G Pluggables for AI Data Center Networks
- Nokia's 800G pluggables for data center interconnect cut power consumption 65-70% versus embedded optics
- Power savings are a key draw for constrained operators handling AI traffic between core and edge
- Vertical integration at its California indium phosphide fab yields ZR/ZR+ edge in transmission distance and supply chain control
Full summary
Jon Baldry from Nokia presents the company's 800 gig pluggable optical connectivity solutions for data center interconnection at Mobile World Congress, demonstrating through augmented reality how the technology reduces power consumption by 65-70% compared to embedded optics—a critical advantage for power-constrained data center operators managing AI-driven traffic between core and edge facilities. Nokia maintains its competitive edge in the 800 gig ZR and ZR plus optics market through vertical integration, manufacturing pluggables, subsystems, and optical chips at its indium phosphide fabrication facility in California, enabling superior transmission distances and supply chain control.
90% of Telcos See Revenue Gains from AI
- NVIDIA's fourth annual AI report finds 90% of telcos say AI is already raising revenue or cutting costs
- Planned AI spending jumped from 65% in 2024 to 89% in 2025
- 77% expect AI-native networks before 6G, targeting spectral and energy efficiency, performance, and personalization
Full summary
Chris Penrose, Global Head Business Development - Telco at NVIDIA, presents findings from NVIDIA's fourth annual AI report showing that 90% of telecom respondents report AI is already increasing revenues or reducing costs, with planned AI spending jumping from 65% in 2024 to 89% in 2025. Network transformation through AI has become the top use case, with 77% of respondents expecting to implement AI-native networks before 6G arrives, focusing on spectral efficiency, energy efficiency, performance improvements, and service personalization.
1Finity Ultra Optical System OLS Built for the AI Era
- 1Finity (Fujitsu) launches L1000 and L2000 ultra optical systems targeted at the AI era
- Compact modular designs span DCI, metro, long-haul, and edge access ROADM applications
- Positioned on lower cost per bit per kilometer and network sustainability
Full summary
Joe Mocerino, Principal Solutions Architect at 1Finity, a Fujitsu company, presents the newly launched L1000 and L2000 ultra optical systems designed as a fundamental foundation for the AI era, addressing high-capacity traffic demands, lower cost per bit per kilometer, and network sustainability through compact modular designs supporting DCI, metro, long-haul, and edge access ROADM applications.
Unpacking the New OCI MSA Optical Standard
- New OCI MSA optical standard targets scale-up networks
- Uses wavelength division with four colors of light per direction instead of serial PAM4 lanes
- Reduces latency, power, and cost versus high-speed serial approaches
Full summary
Near Margalit, GM and VP of Optical Systems Division at Broadcom, presents the newly released OCI MSA optical standard for scale-up networks at OFC26, which reduces latency, power, and cost by utilizing wavelength division technology with four colors of light in each direction instead of high-speed serial PAM 4 lanes.
Broadcom's 400G/Lane Taurus Optical DSP
- Taurus is the industry's first 400 Gbps-per-lambda PAM4 optical DSP
- Enables low-cost, low-power 1.6T transceivers and a path to 3.2T transceivers
- Targets 200 terabit switch networking systems for AI infrastructure
Full summary
Gang Qiu, Product Line Manager, Marketing at Broadcom, presents Taurus, the industry's first 400 gigabit per lambda PAM4 optical DSP designed to help customers scale AI infrastructure at OFC 2026. The solution enables next-generation low-cost, low-power 1.6 terabit transceivers and paves the path to 3.2 terabit transceivers in 200 terabit switch networking systems.
GPU Cluster Connectivity
- Ciena Nitro linear retimer driver targets active copper cables for GPU cluster interconnects
- Vesta 200 is a 6.4T CPX optical engine for high-capacity AI networks
- Positioned as flexible, power-efficient connectivity options
Full summary
Helen Xenos, Senior Director of Portfolio Marketing at Ciena, presents the company's innovations for GPU cluster interconnects at OFC, including the Nitro linear retimer driver for active copper cables and the Vesta 200 6.4T CPX optical engine. These solutions provide flexible, power-efficient connectivity options to help customers build high-capacity networks for modern AI infrastructure.
Ethernet's Critical Role for AI Infrastructure
- AI-driven demand is straining supply chains for chips, power, and fiber
- Ethernet Alliance argues Ethernet's reliability track record makes it the only viable AI connectivity solution
Full summary
Peter Jones, Chair of the Ethernet Alliance, presents at OFC 2026 on how Ethernet technology must address unprecedented AI-driven infrastructure demands that are straining supply chains for chips, power, and fiber. He emphasizes that despite these immense challenges, Ethernet's proven track record of reliability makes it the only viable solution for enabling the industry's AI connectivity requirements.
AI, Security & New Revenue Opportunities
- AI continues to dominate the industry in 2026, but power consumption is the binding constraint
- Digital sovereignty and security advance, including post-quantum cryptography
- Service providers evolve beyond infrastructure into digital platforms
Full summary
Stephan Rettenberger, SVP of Marketing and Corporate Communications at Adtran discusses AI's continued industry dominance constrained by power consumption; digital sovereignty and security advances including post-quantum cryptography; and service providers evolving beyond infrastructure into digital platforms.
Data Center Power, Water & AI Infrastructure Trends
- Power suppliers and data center developers deepen partnerships, with collocated generation and SMRs becoming more prevalent
- GPU cooling demands in warmer regions drive significant water management innovation
- Neocloud providers offer distributed AI-as-a-service in the 20-80 MW range
- Integrated campus simulations and AI-enabled operations become standard practice
Full summary
Jenn Cahill, Assistant VP of Campus Infrastructure Integration at Black & Veatch, projects that 2026 will bring increased partnerships between power suppliers and data center developers, with collocated generation and sustainable options like SMRs becoming more prevalent, alongside significant water management innovation driven by GPU cooling demands in warmer regions. She highlights the rise of neocloud providers offering distributed AI-as-a-service in the 20-80 megawatt range, and emphasizes that data-driven decision-making through integrated campus simulations and AI-enabled operations will become standard practice across the industry.
Ethernet Dominance, Multi-Site Clusters & Co-Packaged Optics
- Ethernet becomes the standard for scale-up networking
- AI clusters expand across multiple data centers with Ethernet as the interconnecting fabric
- Co-packaged optics go mainstream on the back of 70% power efficiency gains
Full summary
Hasan Siraj, Head of Software Products/Ecosystem at Broadcom, presents three networking predictions for 2026 centered on AI infrastructure, including Ethernet becoming the standard for scale-up networking, AI clusters expanding across multiple data centers with Ethernet as the interconnecting fabric, and co-packaged optics going mainstream due to 70% power efficiency gains. Broadcom is investing heavily in these technologies while collaborating with customers and partners to enable these infrastructure transitions.
OIF on 448G, Coherent Lite, CMIS & Power Efficiency
- The industry pushes toward 400/448 Gbps implementation while tackling reliability, power efficiency, and reach
- Coherent lite emerges as a developing market segment
- CMIS becomes critical firmware infrastructure for AI applications
Full summary
Nathan Tracy, President of OIF, discusses 2026 predictions focusing on the industry's push toward 400/448 Gbps implementation while addressing reliability, power efficiency, and reach challenges, with particular emphasis on coherent lite market development and CMIS as critical firmware infrastructure for AI applications.
When AI Gets Normal (Not Smarter)
- AI becomes "normal" as infrastructure economics, security, and compute politics eclipse model-intelligence gains
- Inference costs force new SaaS pricing models
- AI security shifts from policy to system-design requirements
- Data center power demands make geography and local politics critical factors
Full summary
Eugina Jordan, CEO of YOUnified AI, presents three key predictions for 2026, arguing that AI will become "normal" as fundamental infrastructure challenges around inference economics, security requirements, and compute politics take center stage rather than model intelligence improvements. She advises that inference costs will force new SaaS pricing models, AI security will shift from policy to system design requirements, and data center power demands will make geography and local politics critical factors, concluding that founders should "build boring with right economics in mind because boring always scales."
Building Blocks for AI Processing
- Marvell's new die-to-die interface triples bandwidth density.
- Cuts power consumption by 40-70% versus prior approaches.
- IP block improves interconnect between dies and custom HBM.
Full summary
Mark Kuemerle, VP of Technology and CTO of ASIC Business Unit at Marvell, presented the company's new die-to-die interface technology at the AI Infrastructure Summit, highlighting its ability to triple bandwidth density while reducing power consumption by 40-70%. The innovative IP block enhances interconnect capabilities between dies and custom HBM, marking a major step forward in data center AI system development.
Ethernet Matches InfiniBand for AI Clusters
- Juniper's AI-lab and hyperscaler benchmarks show Ethernet matching InfiniBand performance in AI clusters.
- Leading AI providers pick Ethernet for its broad ecosystem and cost advantages.
- Power efficiency is pursued through liquid cooling and advanced switch designs.
Full summary
Amit Sanyal, Senior Director of Data Center Product Marketing at Juniper, shares how Ethernet performance matches InfiniBand in AI clusters through benchmarks from Juniper's AI lab and major hyperscalers. His analysis shows how leading AI providers select Ethernet for its extensive ecosystem and cost advantages, while highlighting their focus on power efficiency through liquid cooling and advanced switch designs.
AI Data Centers and the Hyperscale Model
- AI infrastructure extends well beyond hyperscalers into multiple markets with distinct power and space constraints.
- Geography shapes AI data center requirements, not a single hyperscale template.
- Nokia's approach centers on partnerships, multivendor management, and comprehensive networking.
Full summary
Mike Bushong, Vice President of Data Center at Nokia, explains that AI infrastructure goes beyond just hyperscalers and spans multiple markets with distinct geographical constraints and requirements for power and space. He outlines how Nokia is addressing these needs through partnerships, multivendor management, and comprehensive networking solutions.
Streamlining Is Key to Faster Fiber Deployment
- Collaborative Synergy streamlines make-ready applications between ISPs and power companies.
- Early planning offsets understaffed utilities and varying requirements for better engineering and construction outcomes.
Full summary
Johannes Maassen, President of Collaborative Synergy, discusses his company's efforts to improve make-ready applications for ISPs and power companies at Fiber Connect 2025. His insights address the challenges of understaffed utility companies and varying application requirements, showing how early planning creates better outcomes in engineering and construction phases.
Solving Multi-Kilowatt AI Chip Power Delivery
- PIVR (Package Integrated Voltage Regulator) moves voltage regulators directly into the XPU package.
- The approach boosts current density performance by up to 2x.
- Power transmission losses drop by as much as 85%, addressing multi-kilowatt chip power delivery.
Full summary
Matt Kim, AVP, Custom Cloud Solutions Business Unit, presents PIVR (Package Integrated Voltage Regulator) technology as a solution to power delivery challenges in the multi-kilowatt chip era where moving voltage regulators directly into the XPU package increases current density performance by up to 2x and reduces power transmission losses by as much as 85%.
SNEAK PEAK on How AI Reshapes Data Center Networks
- AI workloads are pushing data center networks toward flatter architectures with far higher reliability.
- AI training's massive power draw and long runtimes make network failures extremely costly.
- The industry is rethinking traditional Ethernet and splitting designs into distinct front-end and backend networks.
Full summary
Rishi Chugh, VP and GM, Data Center Switching at Marvell, joins Roy Chua, Principal and Founder of AvidThink, to explore how AI workloads are fundamentally changing data center networking by demanding flatter architectures and unprecedented levels of reliability. The discussion covers how AI training's massive power consumption and time requirements make network failures extremely costly, driving the industry to reconsider traditional Ethernet approaches and adopt more complex designs with distinct front-end and backend networks.
Internet Underlay: The Future of Enterprise Connectivity & Multi-Cloud Networks
- Enterprises are turning to internet underlay for connectivity, especially in multi-cloud and edge deployments for IT and AI workloads.
- Colt's strategy centers on streamlined, distributed IP networks.
- Routed optical networking is used to drive cost efficiency and sustainability.
Full summary
Mirko Voltolini, VP Innovation at Colt, examines how enterprises are turning to internet underlay for their connectivity requirements, especially in multi-cloud and edge deployments supporting IT and AI workloads. He outlines a trategy of developing streamlined, distributed IP networks that emphasize cost efficiency and sustainability through routed optical networking technologies.
Power & Efficiency for Next-Gen RAN
- Telcos are upgrading networks with AI capabilities and power efficiency via Intel's Xeon processor families.
- New Xeon processors deliver performance gains and power savings.
- The chips support both networking and AI workloads at network edges and cell sites.
Full summary
Alex Quach, VP and GM of the Wireline and Core Network Division at Intel, explains how telecom providers are upgrading networks with AI capabilities and enhanced power efficiency through the company's Xeon processor families. The new processors deliver significant performance gains and power savings while supporting both networking and AI workloads, particularly at network edges and cell sites.
Connecting the AI Data Center Fabric
- Ciena touts an industry-first single-carrier 1.6 Tb/s coherent solution.
- Ultra-low-power metro DCI uses 400 gig coherent pluggables.
- The Nubis Communications acquisition adds linear redrivers for active copper cables and XT optical engines for co-packaged optics.
Full summary
Helen Xenos, Senior Director, Portfolio Marketing at Ciena, presents the company's comprehensive AI infrastructure solutions at OCP, including their industry-first single carrier 1.6 terabit per second coherent solution and ultra-low power metro DCI with 400 gig coherent pluggables. She also discusses Ciena's recent acquisition of Nubis Communications, which enables linear redrivers for active copper cables and XT optical engines for co-packaged optics applications.
How to Maximize GPU ROI in AI Infrastructure Buildouts
- Cornelis targets weak AI infrastructure ROI across compute cycles, power efficiency, and space utilization.
- Its technology increases GPU utilization to improve the economic model.
- Current 400 gig tech already offers Ultra Ethernet features like credit-based flow control and adaptive routing.
Full summary
Lisa Spelman, CEO of Cornelis Networks, discusses how companies struggle to extract sufficient value from their AI infrastructure investments, particularly regarding compute cycles, power efficiency, and data center space utilization to make their economic models viable. She explains that Cornelis Networks addresses these economic challenges by developing technology that increases GPU utilization and delivers enhanced performance, with their current 400 gig technology already providing ultraethernet compliance features like credit-based flow controls and adaptive routing.
Transforming Data Center for AI with HPE
- HPE positions AI-native networking built from the ground up for AI operational challenges.
- Innovations include liquid-cooled switches and Ultra Ethernet solutions with packet trimming.
- Solutions support OCP-specified OB3 racks, underscoring an open networking commitment.
Full summary
Amit Sanyal, Senior Director, Data Center Product Marketing at HPE, presents the company's AI native networking solutions at OCP, which are designed from the ground up to solve operational challenges and optimize networking performance for AI applications. HPE showcases key innovations including liquid cooled switches, UltraEthernet consortium solutions with packet trimming capabilities, and support for OCP-specified OB3 racks, reinforcing the company's commitment to open networking solutions purpose-built for AI.
100% Heat Capture for High-Power AI Infrastructure
- Iceotope uses dielectric fluids and cold plates to capture 100% of thermal load.
- Cutting cooling energy frees more power for GPUs and compute.
- The approach targets thermal challenges from 1 MW racks and 8 kW chips.
Full summary
Neil Edmunds, VP of Product Management at Iceotope, presents advanced thermal management solutions at OCP 2025 that use dielectric fluids and cold plates to capture 100% of thermal load from high-power AI infrastructure, reducing cooling energy consumption. The Iceotope approach enables data centers to allocate more power to GPUs and compute resources while efficiently managing thermal challenges from 1 megawatt racks and 8 kW chips.
Lumentum Optics Powering AI Infrastructure
- Lumentum tackles network power consumption with optical circuit switches and external laser sources for co-packaged optics.
- Portfolio includes 1.6T partial retimed optics modules.
- Lumentum participates in the OCP OCS working group to align vendors and hyperscalers on common standards.
Full summary
Michael DeMerchant, Sr. Director, Product Line Marketing at Lumentum, explains how the AI infrastructure buildout drives hyperscalers to focus on optical technologies, with Lumentum addressing network power consumption challenges through optical circuit switches, external laser sources for co-packaged optics, and 1.6T partial retimed optics modules. DeMerchant also highlights Lumentum's participation in OCP OCS working group standardization efforts to bring industry vendors and hyperscalers together for common standards that accelerate technology adoption.
Marvell's CPO Switch Demo
- Marvell's CPO reference platform is a liquid-cooled 1 OU system with 16 6.4T silicon-photonics light engines.
- Scaling to 32 units per rack means managing over 36,000 fibers.
- Rack-level success hinges on industry solutions for backplane design and fiber routing.
Full summary
Kishore Atreya, Sr. Director, Platform Product Management at Marvell, presents the company's co-packaged optics reference platform featuring a liquid-cooled 1 OU system with 16 6.4T light engines that integrate silicon photonics technology for electrical-to-optical signal conversion. He addresses the significant deployment challenge of managing over 36,000 fibers when scaling to 32 units per rack, emphasizing the need for industry solutions in backplane design and fiber routing for successful rack-level implementation.
Unlocking Memory Bandwidth
- Marvell's Structura CXL portfolio targets rising memory capacity and compute demands.
- Structura A packs 16 ARM cores for deep learning and inference workloads.
- Structura X enables memory expansion, mixing DDR4 and DDR5 with compression algorithms.
Full summary
Khurram Malik, Sr. Director, Product Marketing, CXL at Marvell, presents the company's Structura CXL product portfolio designed to address growing memory capacity and compute power demands. The portfolio includes Structura A with 16 ARM core processors for deep learning and inference workloads, and Structura X for memory expansion solutions that enable hyperscalers to combine DDR4 and DDR5 memory with compression algorithms.
UALink for Scale-Up AI Interconnects
- UALink is pitched as a scale-up fabric with lower power, reduced latency, and efficient data movement.
- It achieves nearly 200 Gbit/s on a 200 Gbit/s link.
- Initial devices are expected in 2026, with widespread data center deployment in 2027.
Full summary
Kurtis Bowman, Chairman of UALink Consortium, presents UALink as a scaleup fabric solution that offers technical advantages including lower power consumption, reduced latency, and efficient data movement achieving nearly 200 Gbits per second on a 200 Gbit per second link. He reports that consortium members have IP available for designing switches and accelerators, with initial device availability expected in 2026 and widespread data center deployment anticipated in 2027.
VIAVI's 800G AI Network Testing
- VIAVI's B3 and M1 appliances generate 800-gigabit traffic with smaller footprint and lower power than traditional methods.
- A Juniper collaboration demonstrates real-time network performance testing.
- VIAVI showcases early Ultra Ethernet Consortium specs, enabling piece-wise testing of emerging layers and 800-gigabit NICs.
Full summary
Kevin Chang, Engineering Director, Hardware Platforms at VIAVI, presents advanced Ethernet test appliances at the OCP Global Summit that provide efficient testing for AI networks through B3 and M1 appliances generating 800-gigabit traffic with smaller footprint and lower power consumption than traditional GPU data center methods. VIAVI collaborates with Juniper to demonstrate real-time network performance testing and showcases early Ultra Ethernet Consortium specifications, enabling piece-wise testing of emerging network layers and 800-gigabit NICs as technologies move from specifications to hardware implementations.
Faster Cadence - Moving to 200G/Lane and Faster
- Acacia launched a silicon photonic Optical Engine family delivering 200G per lane for AI infrastructure.
- A new 1.6T PAM4 DSP marks Acacia's expansion into data center markets.
- Design emphasis on power efficiency and enhanced network performance.
Full summary
Tom Williams, VP, Marketing at Acacia , introduces their new silicon photonic Optical Engine product family designed to support AI infrastructure with 200Gig per lane capabilities across multiple applications. The company's development of a 1.6T PAM4 DSP demonstrates their expansion into data center markets while focusing on power efficiency and enhanced network performance.
Next-Gen Optical from VCSELs to CPO
- Broadcom debuts 200Gbps per lane VCSEL technology plus production-ready EML lasers for AI networks.
- Its co-packaged optics reduce power usage by 70%.
- The Tomahawk 5 Bailly 51T Ethernet switch connects to Thor 2 NIC for XPU-to-switch optical connectivity.
Full summary
Manish Mehta, VP Marketing, Optical Systems Division of Broadcom, presents the company's optical interconnect advances for AI networks at OFC, including their groundbreaking 200Gbps per lane VCSEL technology and production-ready EML laser solutions. The presentation showcases how Broadcom's co-packaged optics reduce power usage by 70% while enabling enhanced integration through the Tomahawk 5 Bailly 51T Ethernet switch connected to Thor 2 NIC and XPU-to-switch optical connectivity solutions.
Copper Replacing Fiber in AI Data Centers?
- Liquid cooling has shortened data center connectivity needs to 1-2 meters, enabling copper solutions.
- Credo's 800G zero-flap Active Electrical Cables are replacing optical links in hyperscale environments.
- Copper's power savings can be redirected to GPU operations.
Full summary
Don Barnetson, SVP & Head of Product at Credo, explains how liquid cooling has shortened data center connectivity needs to 1-2 meters, enabling new copper-based solutions. His company's 800G zero flap Active Electrical Cables are replacing optical alternatives in hyperscale environments, delivering better reliability and significant power savings that can be redirected to GPU operations.
800G Testing Challenges & Solutions for High-Speed Networks
- EXFO's Anabel Alarcon covers 800G advances and interconnection solutions at OFC 2025.
- 800ZR technology enables transmission up to 120 kilometers for metro and data center links.
- Thorough testing and thermal management grow critical as AI drives cooling requirements.
Full summary
Anabel Alarcon, Product Line Manager at EXFO and High-Speed Tech Lead at Ethernet Alliance, discusses the advancement of 800G technologies and interconnection solutions at the OFC 2025 booth. She explains how 800 ZR technology enables data transmission up to 120 kilometers for metro internet and data center connections, while highlighting the importance of thorough testing and thermal management as AI applications drive cooling requirements.
Next-Gen Switching: Co-Packaged Optics vs. Co-Packaged Copper
- Marvell showcases a co-packaged copper switch-tray design using substrate-mounted flyover wires at 224G.
- The liquid-cooled solution enables direct XPU ASIC to CPO engine connections.
- Supports large-scale multi-rack clusters up to 2,000 nodes with a single switching layer.
Full summary
George Hervey, Principal Architect at Marvell, showcases an innovative co-packaged copper design for switch trays that enhances passive copper performance through substrate-mounted flyover wires and 224G compatibility. The liquid-cooled solution enables direct XPU ASIC to CPO engine connections and supports large-scale multi-rack compute clusters up to 2,000 nodes with a single switching layer.
OIF's Energy-Efficient Interfaces for AI/ML
- OIF President Nathan Tracy demos energy-efficient ELSFP modules and inspection equipment.
- Covers architectural implementations for AI/ML data centers and high-density networking.
- OIF plans a 400Gbit workshop in Santa Clara.
Full summary
Nathan Tracy, President of OIF, demonstrates energy-efficient interface solutions including ELSFP modules and inspection equipment at OFC 2025 in San Francisco. His presentation covers architectural implementations for AI/ML data centers and highlights OIF's commitment to advancing high-density networking solutions, with plans for a 400Gbit workshop in Santa Clara.
Wafer-Scale CPO + Wafer-Scale AI Computing
- Ranovus CEO Hamid Arabzadeh details wafer-scale co-package optics spanning multiple GPUs.
- Delivers significantly higher capacity than traditional solutions while addressing power and cost.
- Developed with Cerebras and backed by a DARPA contract.
Full summary
Hamid Arabzadeh, CEO of Ranovus, details their wafer-scale co-package optics technology that enables optical interfaces across multiple GPUs, delivering significantly higher capacity than traditional solutions. The technology, developed through collaboration with Cerebras and backed by a DARPA contract, addresses key power and cost challenges while providing superior interconnect capabilities for AI compute platforms.
Connecting Next-Gen AI | 800G to 3.2T
- Semtech's Raza Khan showcases power-efficient 800G LPO with ACC interconnects.
- Roadmap advances toward 1.6T and 3.2T capabilities.
- Also features 50G PON and wireless infrastructure products for 6G applications.
Full summary
Raza Khan, Product Marketing Director at Semtech, showcases the company's power-efficient AI interconnect solutions including 800G LPO with ACC interconnects at OFC. The presentation highlights Semtech's advancement toward 1.6T and 3.2T capabilities while featuring 50G PON and wireless infrastructure products for 6G applications.
Rethinking Connectors with 2D Interfaces for the 448G Era
- Marvell is researching high-density 2D connector designs to support 448G signaling.
- The goal is lossless electrical channels for next-generation AI data centers.
- Design priorities center on signal integrity optimization and thermal efficiency.
Full summary
Matt Traverso, Distinguished Engineer at Marvell, leads research on high-density 2D connector designs to enhance 448G signaling capabilities for AI and large-scale computing. His work prioritizes signal integrity optimization and thermal efficiency to create lossless electrical channels for next-generation data centers.
ling Bandwidth from 228 to 448G
- Meta is targeting a scale from 228 to 448 Gbits per second for AI networks.
- Power and space constraints are key limits on next-gen bandwidth.
- Modulation formats and interconnect components need industry alignment to advance.
Full summary
Xu Wang, Hardware Engineer at Meta, examines the growing network bandwidth demands for AI systems, focusing on scaling from 228 to 448 Gbits per second while managing power and space constraints. He outlines technical hurdles in modulation formats and interconnect components, emphasizing the need for industry alignment to support next-generation AI infrastructure development.
448G Connectivity: Key Requirements & Solutions
- Microsoft frames the shift from 224G to 448G around power budget and retimer challenges.
- Requirements include roughly 40 dB reach specifications alongside power and cost targets.
- Both copper and optical implementation approaches are on the table.
Full summary
Ashwin Gumaste, Principal AI Systems Architect at Microsoft, examines the industry's shift from 224G to 448G connectivity, focusing on power budget constraints and retimer solution challenges. He explores technical requirements including 40 dB reach specifications, power optimization targets, and cost considerations while discussing both copper and optical implementation approaches.
Gearbox-Free Electrical-Optical Co-Design for 448G Era
- Nubis argues for tight electrical-to-optical integration in the 448G era.
- Its approach eliminates power-consuming gearboxes.
- Optics can be driven directly from SerDes.
Full summary
Peter Winzer, Founder & CTO of Nubis Communications, emphasizes the importance of electrical-to-optical integration at the OIF 448G workshop, focusing on solutions that operate without power-consuming gearboxes and can be directly driven from SerDes.
Why 400G Systems Demand a New Era of Industry-Wide Collaboration
- The shift to 400G systems demands unprecedented collaboration across equipment vendors, component makers, and chip designers.
- OIF is coordinating industry-wide evaluation of circuit designs and modulation approaches.
- Power and reliability hurdles remain the main obstacles to 400G deployment.
Full summary
Cathy Liu, SerDes Architect at OIF, explains how the shift to 400G systems demands unprecedented collaboration between equipment vendors, component makers, and chip designers to overcome bandwidth constraints. She describes how OIF is coordinating industry-wide efforts to evaluate circuit designs and modulation approaches that will enable successful 400G deployment despite power and reliability hurdles.
Energy-Efficient Solutions for AI Infrastructure
- OIF's Energy Efficient Interfaces group builds on its 2020 co-packaging work with hyperscalers.
- It advances transmit linear receiver solutions and high-density connectors.
- Power consumption and reliability are the critical challenges in AI training environments.
Full summary
Jeff Hutchins, Director of the CTO Office at OIF, leads the organization's Energy Efficient Interfaces group, which builds upon their 2020 co-packaging work with hyperscalers. The group advances multiple energy-saving technologies, including transmit linear receiver solutions and high-density connectors, while addressing critical power consumption and reliability challenges in AI training environments.
5G Standalone & Open RAN in 2025
- Three 2025 wireless predictions: zero-trust security, 5G standalone expansion, and more Open RAN.
- These shifts help operators maximize network investments and advance energy efficiency goals.
Full summary
Bill McKenney, Marketing Director at Analog Devices, shares three key wireless communications predictions for 2025, including zero-trust security adoption, expanded 5G standalone networks, and increased Open RAN deployment. His analysis highlights how these changes will help operators maximize their network investments while advancing energy efficiency goals across the telecommunications sector.
3 Critical Network Trends Shaping 2025
- Fiber availability and route diversity become central to network resilience.
- AI advancement drives new power-infrastructure requirements.
- Sustainability reporting becomes mandatory for network operators.
Full summary
Mattias Fridström, Chief Evangelist at Arelion, outlines three essential predictions for 2025, focusing on fiber availability for network resilience, power infrastructure requirements for AI advancement, and mandatory sustainability reporting. His analysis highlights how network operators must adapt to increasing demands for route diversity while balancing power resources and environmental responsibilities.
AI Infrastructure Predictions That Will Shape 2025
- AI shifts from training to inference in 2025.
- AI models increasingly deploy at network edges.
- Arrcus embeds security into routing and switching while optimizing channels for power and cost efficiency.
Full summary
Shekar Ayyar, CEO of Arrcus, outlines three major AI infrastructure predictions for 2025, including a shift from training to inference and the deployment of AI models at network edges. His analysis highlights how Arrcus is developing intelligent networks with optimized channels for power and cost efficiency, while integrating security features directly into routing and switching infrastructure.
AI Networks Drive Enterprise Sustainability Goals
- Sustainability becomes a strategic necessity rather than a corporate checkbox.
- AI-driven networks with Wi-Fi 7, IoT protocols, and private 5G improve energy efficiency and resource management.
- Applies across hospitality, education, healthcare, and corporate environments.
Full summary
Mittal Parekh, Senior Director of Ruckus Networks Business Unit at CommScope, outlines how network technologies will drive enterprise sustainability in 2025, positioning sustainability as a strategic necessity rather than a corporate checkbox. He illustrates how AI-driven networks, combined with Wi-Fi 7, IoT protocols, and private 5G, will enhance energy efficiency and resource management across various enterprise settings including hospitality, education, healthcare, and corporate environments.
Sustainability, AI, Wi-Fi 7 & Security
- Four major trends: sustainability, AI influence, wireless networks, and Open Standards adoption.
- Passive optical LANs and Wi-Fi 7 improve power efficiency, speed, and security.
Full summary
Tim Dyer, Vice President of Market Development at CommScope, outlines four major trends in the W sector including sustainability, AI influence, wireless networks, and Open Standards adoption. He highlights how passive optical LANs and Wi-Fi 7 technologies are advancing network capabilities through improved power efficiency, increased speeds, and enhanced security measures.
Optical DSP Evolution - Pluggables and More
- 1.6T transceivers and power-efficient 3nm DSP devices emerge in 2025.
- Optical link reliability becomes essential for AI clusters.
- The optical ecosystem expands beyond traditional suppliers to meet growing demand.
Full summary
Chris Collins, VP of Sales & Marketing for Optical DSPs at Credo, outlines predictions for 2025 including the emergence of 1.6T transceivers and power-efficient 3nm DSP devices. He highlights how optical link reliability will be essential for AI clusters while the optical ecosystem expands beyond traditional suppliers to support growing market needs.
Ethernet, Standards & Copper's Comeback
- InfiniBand's role shrinks as Ethernet becomes the standard for scale-out networks.
- Proprietary interfaces align with 224G IEEE Ethernet standards.
- Liquid cooling revives copper for NIC-to-switch links, offering better power efficiency and reliability at lower cost.
Full summary
Don Barnetson, VP Product - HiWire AECs at Credo, forecasts key networking shifts for 2025, including InfiniBand's reduced role as Ethernet becomes the standard for scale-out networks and proprietary interfaces align with 224G IEEE Ethernet standards. His analysis highlights how liquid cooling's space advantages will drive renewed interest in copper connectivity for NIC-to-switch connections, offering better power efficiency and reliability at lower costs.
3 Major Data Center Predictions
- AI infrastructure demand, power constraints, and specialized facility shortages dominate 2025.
- These obstacles drive innovations in energy solutions and cooling systems.
- Resulting design choices could influence data centers through 2050.
Full summary
Yuval Bachar, Co-founder and CEO of ECL, forecasts major data center challenges for 2025, focusing on unprecedented AI infrastructure demands, power constraints, and specialized facility shortages. His analysis suggests these obstacles will drive innovations in energy solutions and cooling systems that will influence data center design through 2050.
3 Enterprise Wireless Network Trends
- Private network adoption expands across industries like energy and healthcare.
- Networks support multiple wireless technologies for indoor and outdoor connectivity.
- Simplified deployment and enhanced partnerships make private networks more attainable.
Full summary
Kelly Burroughs, Enterprise Solutions Director at iBwave, forecasts that enterprise wireless networks in 2025 will see expanded private network adoption across industries like energy and healthcare, while supporting multiple wireless technologies for indoor and outdoor connectivity. She highlights that simplified deployment methods and enhanced partnerships will make private networks more attainable for businesses.
Private Networks, Edge Computing & AI Evolution
- Three 2025 initiatives: private networking, edge computing, and AI for operations.
- AI power consumption in data centers is a key challenge.
- Distributed computing models become necessary.
Full summary
Marc Cohn outlines three major technology initiatives for 2025: private networking for industrial and enterprise applications, edge computing for enhanced service delivery, and AI for operational improvements. His analysis emphasizes how these technologies will shape network infrastructure development while addressing key challenges like AI power consumption in data centers and the need for distributed computing models.
Data Center Tech Evolution in 2025
- Hybrid cloud adoption increases in 2025.
- AI deployments shift from InfiniBand to Ethernet.
- Energy optimization comes via efficient optical modules, liquid cooling, and AI-driven facilities management.
Full summary
Amit Sanyal, Senior Director of Data Center Product Marketing at Juniper, outlines key networking trends for 2025, including increased hybrid cloud adoption and a shift from InfiniBand to Ethernet for AI deployments. He emphasizes the importance of energy optimization through efficient optical modules, liquid cooling solutions, and AI-driven facilities management systems in data centers.
AI Automation, Sustainability & 800G
- Intent-based closed-loop automation advances.
- AI-driven sustainability initiatives grow.
- 800-gig networking sees widespread adoption across cloud, telecom, and data center environments.
Full summary
Jai Thattil, Head of Industry & Sustainability Marketing at Juniper, highlights three key networking trends including intent-based closed-loop automation, AI-driven sustainability initiatives, and widespread 800-gig networking adoption. These trends reflect the industry's focus on enhancing network operations through AI capabilities while addressing environmental concerns and meeting growing bandwidth demands across cloud, telecom, and data center environments.
AI, Edge Computing & Nuclear-Power
- Edge computing and AI lead substantial IT infrastructure expansion across cloud, enterprise, and satellite networks.
- Small modular nuclear reactors emerge to power cloud infrastructure.
- Encryption and decentralized identity management are critical to securing evolving systems.
Full summary
Ed Fox, CTO of MetTel, forecasts substantial IT infrastructure expansion in 2025, with edge computing and AI leading the advancement across cloud, enterprise, and satellite networks. He anticipates the emergence of small modular nuclear reactors for powering cloud infrastructure, while highlighting the critical role of encryption and decentralized identity management in securing these evolving systems.
3 Game-Changing Tech Trends for 2025: AI, Cybersecurity & Sustainability | Business Leaders
- Generative AI becomes a business standard, backed by $2 trillion in global investment.
- Cybersecurity evolves into an AI-driven challenge.
- NTT Data cites a $3.3 billion Toyota AI collaboration and a pledge for carbon-neutral data centers by 2030.
Full summary
Tammy Soares, President of Launch by NTT Data, highlights three major tech trends for 2025: generative AI becoming a business standard with $2 trillion in global investment, cybersecurity evolving into an AI-driven challenge, and the intersection of innovation with sustainability goals. Through strategic partnerships and investments, NTT Data is advancing these priorities with initiatives like their $3.3 billion Toyota collaboration for AI platforms and their commitment to achieve carbon neutrality in data centers by 2030.
Linear Pluggable Optics Set to Redefine Data Centers
- Linear optics adoption grows for high-speed interfaces.
- Retiming driver advancements and enhanced link training algorithms arrive.
- These changes improve latency, power efficiency, and cost across networking and compute applications.
Full summary
Armond Hairapetian, Founder and CTO of Terasignal, identifies three major networking developments for 2025, including linear optics adoption for high-speed interfaces, retiming driver advancements, and enhanced link training algorithms. His analysis highlights how these changes will lead to significant improvements in latency, power efficiency, and cost reduction across networking and compute applications.
Power, Cooling & the Future of Data Center Design
- Power, cooling, and space constraints increasingly drive data center network design.
- 1.6 terabit deployments with hybrid cooling anticipated by 2026.
- Scale-across connectivity between distributed data centers is a key trend for the coming year.
Full summary
Arihant Jain, Manager, Systems Engineering at Arista Networks, examines scale-out networking architectures and interconnect technologies while highlighting how power, cooling, and space constraints are increasingly driving data center network design decisions alongside traditional requirements. He emphasizes Ethernet's expanding role across scale-up and scale-out applications, anticipates 1.6 terabit deployments with hybrid cooling solutions by 2026, and identifies scale-across connectivity between geographically distributed data centers as a key deployment trend for the coming year.
Ethernet for AI: Defining Infrastructure Standards
- Ethernet Alliance conference convenes decision makers from Meta, Microsoft, and Hewlett Packard.
- Focus is on the convergence of power and data for AI and quantum computing infrastructure.
- Interconnect industry learns required innovations, codes, and standards directly from technology providers.
Full summary
David B. Kiddoo, Executive Director / CEO of CCCA and IWCS, explains his organizations' sponsorship of the Ethernet Alliance conference that brings together decision makers from Meta, Microsoft, and Hewlett Packard to address Ethernet for AI and the convergence of power and data supporting AI and quantum computing infrastructure. The conference enables the interconnect and network infrastructure industry to learn directly from technology providers about required innovations, necessary codes and standards, and how to allocate limited resources effectively to advance AI and quantum computing technologies.
Building Next-Gen Ethernet: Collaboration, Innovation & Industry Challenges
- Next-gen optical modules need doubled improvements in signal integrity, power, cost, and manufacturing yield.
- UltraEthernet standards work demands cross-industry collaboration.
- The timeline for these next-generation innovations remains uncertain.
Full summary
Adee Ran, Architect of Ethernet at Cisco Systems, discusses his standards work with IT and UltraEthernet, emphasizing the collaborative effort required across industries to develop next-generation networking technologies. He explains that achieving mature next-generation technology requires inventing numerous components, such as optical modules with doubled improvements in signal integrity, power consumption, cost, and manufacturing yield, though the timeline for these innovations remains uncertain.
AI Networking at 1.6T & 3.2T: Performance vs Power Trade-offs
- 1.6T Ethernet deployment has begun while hyperscalers plan for 3.2T Ethernet.
- 400G signaling is advancing from lab to product amid performance, power, reliability, and supply chain trade-offs.
- Keysight works with customers on testing requirements at these speeds.
Full summary
Hadrien Louchet, Product Planner at Keysight Technologies, presents at a technology exploration forum on AI networking's future, noting that while 1.6T Ethernet deployment has begun and hyperscalers are planning for 3.2T Ethernet, the industry must navigate critical trade-offs between performance, power efficiency, reliability, and supply chain security as 400G signaling advances from lab to product. Louchet explains that Keysight, as a test and measurement expert, works with customers to understand testing requirements at these speeds and prepares products to support large-scale deployment of next-generation AI networking infrastructure.
Domain-Specific AI for Networking
- The Linux Foundation's Arpit Joshipura sees domain-specific AI as the next development in verticals like telecom and energy.
- Linux Foundation Networking collaborates with projects to anonymize data for model training.
Full summary
Arpit Joshipura, GM Networking + Edge IOT at the Linux Foundation, discusses the emergence of domain-specific AI as the next major development in enterprise verticals like telecommunications and energy. He highlights the Linux Foundation Networking's collaboration with projects to anonymize data for model training.
The Power of AI in Campus NaaS
- Juniper's Gordon Mackintosh highlights an OpEx-based networking model that eliminates upfront investment.
- Juniper's partner-first, AI-native approach helps partners speed deployment and achieve scale via the cloud.
Full summary
Gordon Mackintosh, Group Vice President of Juniper's Partner Organization, highlights the flexibility of consuming networking in an OpEx-based model, eliminating upfront investments. He emphasizes how Juniper's partner-first approach to NaaS, combined with their AI-native solutions, allows partners to deliver excellent customer experiences, speed up deployment, and achieve scale through the cloud.
BT is Building a Global Fabric for NaaS
- BT delivers carrier NaaS via its Global Fabric product, consolidating multiple services into a single platform.
- BT's redesigned network prioritizes security, programmability, and sustainability while aligning with cloud providers' edges for global connectivity.
Full summary
Colin Bannon, CTO of BT, outlines the company's implementation of carrier Network-as-a-Service (NaaS) through their Global Fabric product, which consolidates multiple services into a single platform. BT's redesigned network prioritizes security, programmability, and sustainability, while aligning with cloud providers' edges to ensure global connectivity.
Open Networks are the Key
- Condor Technologies advocates open networks for sustainable growth and cost-effective multi-operator sharing.
- Its Host Network Portal integrates with Nokia's Altiplano using Broadband Forum standard protocols.
Full summary
Alberto Patron, CEO of Condor Technologies, advocates for open networks at Fiber Connect, emphasizing their importance for sustainable fiber network growth and cost-effective multi-operator sharing. Condor Technologies offers solutions for operators to open and manage networks for various ISPs, showcasing their Host Network Portal solution integrated with Nokia's Altiplano using standard protocols defined by the Broadband Forum.
Can it be done? Powering a Million XPUs
- Marvell's Loi Nguyen warns future gigawatt-class data centers will need energy comparable to major cities.
- Multi-site 400k+ node clusters will be linked by high-speed optics like Marvell's DCI Optics ZR and ZR+.
Full summary
Loi Nguyen, EVP and GM of Optical, highlights the enormous power demands of modern data centers and AI clusters, with future "gigawatt-class" facilities requiring energy comparable to major cities. Nguyen discusses the need for multi-site 400k+ node clusters connected by high-speed optics like Marvell's DCI Optics ZR and ZR+.
The Future of Co-Packaged Optics Looks Bright
- Marvell's Nick Kucharewski positions co-packaged optics for scale-up networking with lower power and latency.
- Marvell's light engine provides up to 51 Tbps of bandwidth for scale-up switches.
Full summary
Nick Kucharewski, SVP and GM of Network Switching, highlights co-packaged optics (CPO) as a promising technology for scale-up networking, integrating fiber links directly into product packages for higher speed signals with lower power consumption and latency. Marvell's light engine provides to 51 Tbps bandwidth for scale-up switches and supports reliable large-scale deployment.
Introducing "Coherent-Lite" for Data Center Connectivity
- Marvell's Achyut Shah introduces coherent-lite for intermediate-reach data center connectivity.
- It combines coherent modulation, O-band optics, and a re-architected DSP for links from 2 km to 20 km.
Full summary
Achyut Shah, SVP and GM of Connectivity, introduces the company's new "coherent-lite" technology, addressing intermediate-reach connectivity needs in data centers. This solution combines coherent modulation with O-band optics and a re-architected DSP to provide lower latency, cost, and power consumption for links ranging from 2 km to 20 km.
PAM Technology: Powering Data Centers Today and Tomorrow
- Marvell's Achyut Shah highlights PAM technology's dominance in data centers for power efficiency and low latency.
- He predicts 400 Gig per lane becomes the high-volume data center technology within a couple of years.
Full summary
Achyut Shah, SVP and GM of Connectivity, discusses the current dominance of PAM technology in data centers, highlighting its advantages in power efficiency, low latency, and high TCO. Shah predicts that 400 Gig per lane will become the high-volume technology in data centers within a couple of years, with ongoing R&D efforts to extend PAM's capabilities even further.
Custom Silicon is the Driver of Optimized AI Infrastructure
- Marvell's Raghib Hussain sees optimized AI infrastructure built on custom silicon for data centers.
- Partnerships with Amazon and Meta show customization delivering power and cost savings at scale.
Full summary
Raghib Hussain, President of Products and Technologies, highlights the growing trend of developing optimized AI infrastructure using custom silicon for data centers. Hussain discusses Marvell's partnerships with major tech giants like Amazon and Meta, emphasizing the value of customization in achieving power and cost savings at scale.
Optimizing the Stack for AI Models and Infrastructure
- Marvell's Noam Mizrahi contrasts high-performance AI training with cost-effective inference, requiring customization.
- Hyperscalers must optimize everything from accelerator chips to power systems for effective scaling.
Full summary
Noam Mizrahi, EVP and Corporate CTO, highlights the contrasting demands in AI between high-performance training and cost-effective inference, emphasizing the necessity for customization to meet diverse requirements. He anticipates that hyperscalers will need to optimize all aspects of their platforms, from accelerator chips to power systems, to achieve effective scaling in the future.
Marvell's Breakthrough High Bandwidth Memory Architecture
- Marvell's Will Chu announces partnerships with major HBM providers for custom high bandwidth memory.
- Custom HBM cuts power consumption, boosts memory capacity, and improves TCO for AI data centers.
Full summary
Will Chu, SVP and GM of Custom Compute and Storage BU, announces a partnership with major HBM (high bandwidth memory) providers to deliver custom HBM solutions for next-generation AI data centers. This enables significant reductions in power consumption, increased functionality, and improved memory capacity, resulting in better performance and enhanced TCO for Marvell's customers.
Satellites Will Soon Transform the Mobile Industry
- EdgeQ's Vinay Ravuri sees satellites, especially LEO, adopting 4G and 5G protocols for direct phone communication.
- EdgeQ's software-defined physical layer and low-power design make satellites cheaper and more efficient despite radiation-hardening challenges.
Full summary
Vinay Ravuri, CEO of EdgeQ, discusses the satellite industry's shift towards adopting traditional cellular protocols like 4G or 5G, particularly in Low Earth Orbit (LEO) deployments, and the potential of using regular phones for direct communication. Despite challenges like the need for radiation-hardened chips and long revenue cycles, EdgeQ's software-defined physical layer and low-power solution can make satellites less expensive and more efficient.
Harnessing AI for the Network Edge
- Intel's Cristina Rodriguez stresses networks must be ready to support AI with existing technology.
- Intel shipped its 4th-gen Xeon with vRAN Boost in 2023, plans Granite Rapids D in 2024, and offers a vRAN AI development kit.
Full summary
Cristina Rodriguez, VP Network & Edge Group and GM Wireless Access Network Division at Intel, highlighted the importance of AI in networking and cloud technology, emphasizing that the network must be prepared to support AI capabilities with existing technology. She discussed Intel's role in this, mentioning the launch of the fourth generation of Intel's scalable processor with vRAN boost in 2023, the future generation platform, Granite Rapid D, in 2024, and the early availability of the Intel vRAN AI development kit, all aimed at enhancing performance, power efficiency, and cost-effectiveness while supporting new revenue streams.
Making Every Connection Count Through the Power of AI
- Juniper's Sally Bament showcased an AI-native networking platform that responds to issues in real time.
- AIOps integration into data center operations and full-stack managed services anchor Juniper's strategy.
Full summary
Sally Bament, GVP of Marketing at Juniper, highlighted the company's AI-native networking platform at the Mobile War Congress 2024, which is designed to enhance network connections by collecting data, responding to network issues in real-time, and ensuring secure infrastructure. Bament also emphasized the integration of AI Ops into data center operations and the use of full-stack managed service capabilities to improve business customer experience, which are key to Juniper's strategy and vision.
AI Opportunity at the Edge
- NTT Data's Devin Yaung says AI effectiveness depends on data collected at the edge, closest to the physical world.
- He warns against cloud over-reliance and stresses IoT sensors, the right network, and edge compute for AI.
Full summary
Devin Yaung, SVP of Global Enterprise IoT at NTT Data, highlights the critical role of Edge computing in AI models, emphasizing that AI effectiveness depends on the data collected at the Edge - the layer closest to the physical environment. He warns of companies exhausting processing power on the Edge due to cloud over-reliance, underlining the necessity for IoT sensors for data automation, and concludes that the right network, devices, and Edge compute are vital for AI development.
Harnessing 5G & AI
- Red Hat's OpenShift AI lets businesses customize AI models for diverse 5G-driven needs.
- Red Hat uses AI to automate and reconfigure networks for capacity and sustainable power usage.
Full summary
Ian Hood, Chief Strategist at Red Hat, highlights the integration of 5G and AI in business operations, particularly through Red Hat's OpenShift AI, which enables businesses to customize their AI models for diverse needs. He also mentions the use of AI in automating and reconfiguring networks for capacity and power, with Red Hat's tool, Anable Light Speed, and collaborations with network companies for sustainable power usage in radio and optical networks.
Cracking the Data Center Thermal Challenge with Carbon Nanotubes
- Carbice's Baratunde Cola introduces vertically aligned carbon nanotube thermal interface materials for data center and AI cooling.
- They hold consistent performance through thermal cycles, improving energy efficiency and asset longevity.
Full summary
Baratunde Cola, CEO and Founder of Carbice, introduces the company's advanced thermal interface materials, highlighting their potential to enhance cooling in data centers and AI applications. Carbice's vertically aligned carbon nanotubes maintains consistent performance throughout thermal cycles, addressing key issues in energy efficiency and asset longevity.
The Bright Outlook for Liquid Cooling
- Dell'Oro's Lucas Beran examines the growing importance of data center liquid cooling.
- He covers single-phase and two-phase direct-chip cooling plus immersion cooling for high-density AI workloads.
Full summary
Lucas Beran, Principal Analyst at Dell'Oro Group, examines the growing importance of data center liquid cooling technologies. He outlines advancements in single-phase and two-phase direct chip liquid cooling, as well as immersion cooling, to meet the demands of high-density computing and AI workloads.
Hydrogen Power for Gigawatt Data Centers
- ECL co-founder Yuval Bachar describes off-grid data centers powered by hydrogen fuel cells.
- The zero-emission, water-positive microgrid addresses unreliable grid power amid rising AI-driven demand.
Full summary
Yuval Bachar, co-founder and CEO of ECL, catches the NGI crew pre-OCP to discuss ECL's next-generation sustainable data centers powered by hydrogen. ECL addresses the challenge of unreliable grid power and the growing demand for AI-driven data centers by offering an off-grid microgrid powered by hydrogen fuel cells that is zero-emission and water-positive.
Custom Silicon for Hyperscale Data Centers
- Marvell's Nigel Alvares details custom silicon co-developed with hyperscalers for optimized performance, power, and cost.
- He announces Meta's FB NIC, open-sourced to OCP and built on Marvell's 5-nanometer platform.
Full summary
Nigel Alvares, VP of Global Marketing at Marvell, discusses the company's collaboration with hyperscalers to create custom silicon for optimized performance, power, and cost. He announces Meta's release of the FB NIC, a network interface controller, which Marvell has open-sourced to the OCP community, demonstrating the use of Marvell's 5-nanometer platform for tailored silicon solutions.
All-Optical Circuit Switching for Scalable Data Centers
- Coherent's Liquid Crystal-based optical switch offers low drive voltage, no moving parts, and data-rate agnosticism.
- The platform cuts power 30-40%, with sampling in coming months and commercialization planned next year.
- Coherent cited strong interest from data center operators, hyperscalers, and OEMs for the switch.
Full summary
Sanjai Parthasarathi, CMO of Coherent, presented the company's Liquid Crystal-based Optical switch platform, emphasizing its low drive voltage, absence of moving parts, data rate agnosticism, and power reduction capabilities of 30-40%, making it an attractive solution for power-hungry data centers. Parthasarathi also revealed plans to sample the platform in the coming months, commercialize it next year, and noted strong interest from data center operators, hyperscalers, and OEMs.
800G Linear Receive Optics Connected to 51.2T Switch
- Credo demoed its Dove 850 Linear Receive Optics (LRO), where only the transmit path uses a DSP.
- Placing a DSP solely on the transmit path for retiming and equalization unlocks dramatic power savings.
Full summary
Easwar Sankar, Optical FAE Director at Credo, demos the company's new Linear Receive Optics (LRO) product, Dove 850. In an LRO transceiver, or Active Optical Cable (AOC), only the transmit path from the electrical input to the optical line side output includes a DSP for signal retiming and equalization. This unlocks dramatic power savings.
Big Power Savings with 800G Linear Receive Optics
- Credo CEO Bill Brennan highlighted energy savings from 800G Linear Receive Optics (LRO) for AI clusters.
Full summary
Bill Brennan, President and CEO of Credo, highlights the energy savings possible with 800G Linear Receive Optics (LRO) for AI clusters.
Rethinking Optical Modules for Liquid Immersion
- Dust Photonics' Carmel-8-IMC transmit chip uses no free-space optics, butt-coupling lasers directly to the PIC.
- Attaching fiber directly to the PIC enables transceiver designs usable for both immersion and air cooling.
Full summary
DustPhotonics’ transmit Carmel-8-IMC chip supports an optical assembly with no free-space optics. Inside the chip, lasers are butt-coupled to the PIC and the fiber is also attached directly to the PIC, thus enabling transceiver designs that can be used for both immersion and air cooling. Ronnen Lovinger, CEO of Dust Photonics, shares a demo.
Indium Phosphide (InP) Interconnect for AI Infrastructure
- Infinera addressed the challenge of longer-reach connectivity between AI clusters.
- Infinera's indium phosphide integrated photonic circuit reduces cost and power while improving manufacturability.
Full summary
Robert Shore, SVP of Marketing at Infinera, highlights challenges network operators face in providing longer reach connectivity between AI clusters. Infinera's solution involves the use of indium phosphide, a technology that enables the creation of a single integrated photonic circuit, reducing costs, increasing manufacturability, and lowering power.
Transforming Cloud Interconnects with 800ZR/ZR+
- Lumentum showed 800G ZR+ and 400F high-power ZR+ modules for coherent metro and longhaul communications.
Full summary
Marc Stiller, VP of Product Line Management at Lumentum, shows the latest 800G ZR+ and 400F high power ZR+ modules for coherent communications in metro and longhaul applications.
Transmit-Retimed Optical DSPs Yield Huge Power Savings
- Marvell's Spica Gen2-T is a transmit-only DSP enabling a new Transmit Retimed Optical (TRO) module.
- The TRO module cuts power consumption by 40% while preserving flexibility and scalability for AI clusters.
Full summary
Marvell VP of Product Marketing, Xi Wang, highlighted the rising demand for optical connectivity in AI cluster. Marvell has introduced a new product, Spica Gen2-T, a transmit-only DSP that enables a new type of optical module, the Transmit Retimed Optical (TRO) module, which can reduce power consumption by 40% while ensuring flexibility and scalability.
A New Class of Silicon Photonics for AI Data Centers
- Marvell unveiled a 3D SiPho engine delivering 6.4 Tbps with 200G I/O for AI applications.
- The 3D SiPho Engine offers 2x bandwidth density and 30% lower power per bit versus 100 Gbps interfaces.
Full summary
Loi Nguyen, EVP and GM of Optical at Marvell, unveiled their 3D SiPho engine, a silicon photonics engine that delivers 6.4 Tbps, enhancing speeds for AI applications. Featuring 200G I/O, the 3D SiPho Engine delivers 2x the bandwidth and input/output (I/O) bandwidth density, and 30% lower power per bit versus devices with 100 Gbps electrical and optical interfaces.
Demonstrating Marvell's Latest Portfolio Solutions for AI
- Marvell showcased a 51 Tbps switch, 800G AEC cables, and 800G optics for low-latency AI switching.
- The demo included a liquid-cooled 51 Tbps platform and another using 64 800G OSFP modules.
Full summary
George Hervey, Principal Architect at Marvell, showcased their AI infrastructure solutions: a 51 Tbps switch, 800G AEC cables, and 800G optics for low-latency switching. The demonstration also highlighted a 51 Tbps platform with a liquid cooling option and another 51 Tbps platform using 64 800G OSFP modules.
Massive Upgrades Coming with 200G per Lane Channels
- Marvell unveiled Nova 2, an optical DSP addressing the shift to 200G per lane.
- Nova 2 enables 1.6 Tbps optical transceiver modules for next-generation AI connectivity and compute.
Full summary
Marvell's Vice President of Product Marketing, Xi Wang, unveiled a new product, Nova 2 designed to address the shift to 200G. Nova 2, an optical DSP, facilitates the construction of 1.6 Tbps optical transceiver modules, enhancing connectivity and computing power for next-generation AI.
Demonstrating Marvell's Latest Portfolio Solutions for AI
- Marvell showcased a 51 Tbps switch, 800G AEC cables, and 800G optics for low-latency AI switching.
- The demo included a liquid-cooled 51 Tbps platform and another using 64 800G OSFP modules.
Full summary
George Hervey, Principal Architect at Marvell, showcased their AI infrastructure solutions: a 51 Tbps switch, 800G AEC cables, and 800G optics for low-latency switching. The demonstration also highlighted a 51 Tbps platform with a liquid cooling option and another 51 Tbps platform using 64 800G OSFP modules.
800ZR, 400ZR and OpenZR+ Optics Interoperability
- OIF's demo showed the first public 800ZR multivendor interop across 800ZR, 400ZR, and ZR+ optics.
- Highlights included OpenZR+ over a 1000km multi-span network and 400ZR collocated with 800ZR.
Full summary
OIF hosted its latest pluggable coherent optics demonstration, showcasing the latest advancements in 800ZR, 400ZR and ZR+ optics interoperability. The demo highlights include the first public 800ZR multivendor interop; high transmit power, OpenZR+ over a 1000km, multi-span network; OpenROADM/ITU-T over that same network; and 400ZR collocated on the single-span network with 800ZR. Karl Gass, Physical and Link Layer Working Group Optical Vice Chair from OIF, provides some highlights:
OIF's demos at OFC 2024
- OIF's OFC 2024 interoperability demonstration featured nearly 50 participating companies.
- The event showcased the first multivendor interop with 800ZR modules, aiming to deliver more data at lower power.
Full summary
Mike Klempa, Product Marketing specialist at Alphawave Semi and chair of the OIF's PLL interoperability working group, announces the OIF's interoperability demonstration at OFC 2024 featuring nearly 50 participating companies. The event showcased the first multivendor interoperability demonstration with 800 ZR modules, aiming to deliver more data at lower power.
Spirent's 800G 802.3df Appliance & LPO optics support
- Spirent unveiled the B3 800G Appliance supporting IEEE 802.3df and linear pluggable optics (LPO).
- The B3 showcased 800G interconnects, including DACs, AOCs, ACC, and ZR transceivers, from 11 partners.
Full summary
Steve Rumsby, Senior Director of Architecture & Platform at Spirent, unveiled the Spirent B3 800G Appliance supporting IEEE 802.3df. The B3 Appliance also supports linear pluggable optics (LPO), an interconnect that uses lower power and latency. He also showcased 800G interconnects (DACs, AOCs, ACC and ZR transceivers) from 11 different partners.
Intelligent Re-drivers for Linear Pluggable Optics
- Terasignal introduced a novel re-timer for linear pluggable optics (LPO) that reduces power.
- The chip adds diagnostic features letting end users assess link quality and extract diagnostic information.
Full summary
Armond Hairapetian, Founder and CTO of Terasignal, introduces a novel re-timer for linear plugin optics (LPO). The device not only reduces power but also introduces diagnostic features in the chip, allowing end users to assess link quality and extract vital diagnostic information.
Service Provider Networking Trends in 2024
- Adtran's Ronan Kelly says better inventory alignment will improve cash flow and expand connectivity investment.
- He predicts efficiency gains to cut capital and energy costs, plus interoperability to consolidate multi-vendor networks cheaply.
Full summary
Ronan Kelly, CTO for EMEA at Adtran, predicts three key trends in service provider networking: better inventory alignment will improve cash flow and allow expanded investment in connectivity, improved efficiency to reduce high capital costs and reduce energy and resource consumption, and interoperability to consolidate networks from different vendors at low cost.
Sustainability & 800G Networking Shape Future
- Juniper's Sally Bament predicts sustainability and green networking becoming crucial in 2024.
- She also forecasts the rise of 800G networking and on-premises automation with cloud-based AI operations.
Full summary
Sally Bament, VP of Cloud & Service Provider Marketing at Juniper, predicts that sustainability and "green networking" will become crucial, the rise of 800G networking, and the adoption of on-premises automation with cloud-based AI operations.
5G, AI & Edge Computing in 2024
- NTT's Parm Sandhu predicts increased 5G adoption, including private 5G networks.
- He expects continued growth of software-defined networks and shares views on security, sustainability, and edge computing.
Full summary
Parm Sandhu, VP of Enterprise 5G Products & Services at NTT, predicts increased adoption of 5G technology, including private 5G networks, the continued growth of software-defined networks for more flexible network management, plus shares his thoughts on security, sustainability, and edge computing.
AI, RIC & Sustainability
- Telecom Infra Project's Eugina Jordan predicts telecoms increasing AI/ML use for operations and automated configuration.
- She anticipates RIC adoption for network optimization and a focus on sustainability and TCO reduction.
Full summary
Eugina Jordan, Chief Marketing Officer at Telecom Infra Project, predicts telecoms will increase their AI/ML use for improving operations and customer experiences, and automating network configuration. Eugina anticipates RIC adoption by mobile operators for network optimization and monetization, and a focus on sustainability and TCO reduction through AI/ML and automation.
SASE & SD-WAN Integration: Boosting Efficiency and Security
- Check Point's Nicolas Stricher discusses integrating SASE and SD-WAN for efficiency and security.
- He also presents on sustainability and energy efficiency aligned with the European Green Deal.
Full summary
Nicolas Stricher, Head of Managed Services Engineering, EMEA & Office of the CTO at Check Point, discusses the integration of SASE and SD-WAN solutions for enhanced efficiency and security at the event. Stricher also presents on sustainability and energy efficiency, emphasizing compliance with the European Green Deal and the increasing consumer preference for environmentally conscious companies.
Small Form Factors for High-Speed Ethernet
- Samtec's Thomas Palkert represents the SNEA small form factor group defining connectors and transceivers.
- He stresses balancing faster Ethernet speeds with higher density, lower power, and reduced cost for AI.
Full summary
Thomas Palkert, System Architect at Samtec, represents the small form factor group of SNEA, which defines connectors and transceivers for optical and storage industries. Palkert emphasizes the challenges of density, power, and cost for Ethernet in AI applications, highlighting the need for faster speeds while achieving higher density, lower power consumption, and reduced costs.
Ampere's Vision for Power-Efficient, Large-Scale AI Computing
- Ampere's Sean Varley stresses optimizing AI models for power efficiency alongside large-scale computing needs.
- Ampere's 192-core AmpereOne parts deliver high compute for inference at less than 200 to 400 watts.
Full summary
Sean Varley, VP of Product Marketing at Ampere, underscored the importance of optimizing AI models for power efficiency and the need for large-scale computing in the context of AI evolution. He also unveiled Ampere's strategy to create CPUs capable of efficiently managing inference at reduced cost and power, citing the recent launch of their 192-core Ampere One Parts that can provide high computational power at less than 200 to 400 watts.
Faster Interconnects for AI Data Centers
- Multilane's Hani Daou cites rising compute needs driving investment in 100 terabit AS6 and 200 gigabit per Lambda optics.
- Multilane sees high demand for its 800 gig systems and copper interconnect test solutions serving chip vendors and cloud providers.
Full summary
Hani Daou, Business Development Manager at Multilane, emphasized the need for data center upgrades due to increasing compute power requirements, with investments being directed towards technologies like 100 terabit AS6 and 200 gigabit per Lambda Optics. Multilane, currently experiencing high demand for its 800 gig systems and copper interconnect test solutions, is well-positioned to support semiconductor vendors, cloud service providers, and interconnect vendors with a strategic roadmap for R&D investment and bandwidth scaling.
Optics Leads the Way for Next Gen Data Centers
- OIF's Nathan Tracy says machine learning demands higher bandwidth density, straining power and thermal management.
- OIF is developing 200 Gbps electrical interfaces, a 112 Gbps linear optics spec, and Energy-Efficient Interfaces.
- OIF also works on CMIS management specs and 400ZR, 800ZR, and 1600ZR for next-gen data centers.
Full summary
Nathan Tracy, President of OIF and technologist at TE Connectivity, underscores the growing significance of machine learning and artificial intelligence in data centers, necessitating higher bandwidth and bandwidth density that will impact power consumption and thermal management. To tackle these issues, OIF is developing a 200 Gbps (Gigabit per second) electrical interfaces, a 112 Gbps linear optics specification, and Energy-Efficient Interfaces (EEI), along with working on CMIS, the common management interface specification, and specifications for 400ZR, 800ZR, and 1600ZR, all crucial for next-generation data centers.
New Tech and Services for Access Networks
- ATX Networks introduced super capacitor technology for improved energy storage in broadband networks.
- ATX launched a fiber services organization to address the looming broadband industry labor shortage.
- DOCSIS 4.0 lets cable operators deliver fiber-like services over existing HFC networks.
Full summary
Jay Lee, Chief Technology and Strategy Officer at ATX Networks, discussed the company's new super capacitor technology for improved energy storage in broadband networks and the launch of a fiber services organization to tackle the impending broadband industry shortage. Lee also emphasized the disruptive potential of fiber as the preferred platform for new builds and its role in the development of DOCSIS 4.0 technology, enabling cable operators to offer fiber-like services within their existing HFC networks.
Unleashing Home Tech Potential with Fiber Advancements
- Corning's Kara Mullaley says fiber bandwidth unlocks home use of AI, virtual and augmented reality.
- More efficient information collection and distribution will improve decisions for individuals, communities, and workplaces.
Full summary
Kara Mullaley, Market Development Manager for Carrier Networks at Corning Optical Communications, forecasts that the enhanced bandwidth from fiber technology will allow homeowners to utilize previously inaccessible technologies such as AI, virtual and augmented reality, and sustainability efforts. Mullaley asserts that the improved efficiency in information collection and distribution will result in superior decision-making for individuals, communities, and workplaces.
EdgeQ's multi-node 4G/5G Base Station-on-a-Chip
- EdgeQ's Vinay Ravuri showcases a multi-node 4G/5G Base Station-on-a-Chip targeting cost and power.
- The platform features an open, customizable, production-grade L1 stack.
Full summary
Cost and power are the biggest challenges in cellular infrastructure. In this Mobile World Congress showcase,Vinay Ravuri highlights EdgeQ's multi-node 4G/5G Base Station-on-a-Chip. EdgeQ's platform uniquely features a production-grade L1 stack that is open and customizable.
New Middle Mile: Is Today's Infrastructure Sufficient for the Next Internet?
- NTT says today's infrastructure cannot support the next internet of 3D animations, virtual worlds, AI, and holograms.
- NTT is shifting from electronics to photonics to reduce power consumption and latency for future infrastructure.
Full summary
Are you wondering if today's infrastructure is enough to support the next internet? In this video, Vito Mabruco, Global CMO from NTT, explains why the answer is no and what we need for the next internet. Here are the top three talking points from the video: * Today's infrastructure is not enough to support the next internet, which will require more capacity for 3D animations, virtual worlds, AI, and holograms. * NTT is addressing the need for the next infrastructure by changing the paradigm from electronics to photonics, which they have been using for long-distance transmission and telecommunications. * By replacing electronics with photonics, NTT can reduce power consumption and latency, which will be crucial for delivering the infrastructure of the future. Watch the video to learn more about NTT's vision for building the next infrastructure based on photonics.
Three Key Findings for AI Data Centers
- 650 Group's Alan Weckel says AI server power needs require server base refreshment and growing liquid cooling adoption.
- He likens AI's impact on data center design to OCP's democratization of compute, while remaining skeptical of immersion cooling.
Full summary
Alan Weckel, Founder and Technology Analyst at 650 Group, emphasized the need for server base refreshment to meet AI server power requirements and the growing interest in liquid cooling solutions for server heat management at a recent technology show. He also highlighted the increasing influence of AI on server and data center designs, likening its potential impact to the democratization of compute by the Open Compute Project (OCP) a decade ago, while expressing skepticism towards immersion cooling due to its data center limitations.
AI Efficiency and Scalability
- Ampere's Sean Varley notes the OCP Summit's shift from traditional equipment suppliers to AI as a major data center workload.
- Ampere formed an AI Platform Alliance to unite vendors around open AI acceleration platforms, emphasizing efficiency and sustainability.
Full summary
Sean Varley, VP of Product Marketing at Ampere, addressed the shift in focus at the OCP Summit from traditional equipment suppliers to AI, which is becoming a significant workload in data centers. To tackle the challenges of proprietary technologies in AI deployment, Ampere announced the formation of an AI Platform Alliance, aiming to unite vendors in creating open technology platforms for AI acceleration, emphasizing efficiency and sustainability.
Broadcom's new 25.6T Router Super Chip
- Broadcom unveils the Qumran3D, the industry's first 5-nanometer 25.6 terabit router chip, with up to 66% power savings and 80% space reduction.
- It supports up to 800GE ports using 100Gb/s PAM-4 SerDes and massive on-chip forwarding databases, eliminating companion devices.
Full summary
Henry Wu, Technical Director at Broadcom, unveils the Qumran3D, the industry's first 5-nanometer 25.6 terabit router chip, offering up to 66% power savings and 80% router space reduction. The device, designed to combat cybersecurity threats, supports high-speed, high-density port interfaces up to 800GE leveraging best-in-class 100Gb/s PAM-4 SerDes. The design also enables massive on chip forwarding databases, eliminating the need for companion devices
Power Efficiency of Next Gen CPO Data Center Switches
- Micas Networks presented switches co-packaged with Broadcom optics delivering up to 30% lower power consumption.
- Its co-packaged optics technology interests large hyperscalers seeking data center power savings.
Full summary
Max Simmons, Chief Marketing Officer at Micas Networks, presented at the OCP Global Summit 2023 about the company's new switches co-packaged with optics from Broadcom, which deliver up to 30% lower power consumption. Micas Networks showcased their C-package optics technology, which is of interest to large hyperscalers for power savings in data centers, and is eager to collaborate with these companies for its implementation in the future.
Future Tech Symposium - AI, Quantum, Sustainability
- OCP's Lesya Dymyd highlights the Future Technology Symposium featuring startups, R&D organizations, and academic researchers.
- The 2023 edition ran five technological tracks with live demos; the next event is in Lisbon on April 24 and 25, 2024.
Full summary
Lesya Dymyd, Chair of the Future Technology Symposium at the Open Compute Project, highlighted future technologies presented by startups, R&D organizations, and academic researchers. The 2023 edition featured five technological tracks, live demos and poster sessions. The next event will next take place in Lisbon, Portugal, on April 24 and 25th, 2024.
AI is Open Compute Project's Biggest Challenge
- OCP Board Chair Zaid Kahn says rapid AI development forces a rethink of system design, data center construction, and cooling.
- He emphasizes collective effort and collaboration, explaining the role OCP plays.
Full summary
Zaid Kahn, Board Chair of Open Compute Project (OCP), recognizes the challenges posed by the rapid development of AI, necessitating a rethink in system design, data center construction, and infrastructure cooling. Khan emphasizes the need for collective effort and collaboration to address these issues, and explains the role OCP plays.
Pathways to Optical Innovation for AI Data Centers
- OIF's Nathan Tracy discusses developing 200G electrical and 100G linear optics, CMIS activities, and co-packaging for AI data centers.
- He introduces a new project creating energy-efficient interconnects with interoperable solutions for next-gen AI data centers.
Full summary
Nathan Tracy, President of OIF, recently discussed the organization's future plans, including the development of 200G electrical and 100G linear optics, CMIS activities, and the importance of co-packaging for AI data centers. Tracy also introduced a new project aimed at creating energy-efficient interconnects to deliver industry-standard interoperable solutions for next-generation AI data centers.
What's next for networking in 2023?
- Ekinops' Sylvain Quartier weighs supply chain, energy crisis, cybersecurity, and sustainability pressures on networking.
Full summary
How will supply chain challenges, the energy crisis, cybersecurity, and sustainability impact the networking sector in 2023? Ekinops' Sylvain Quartier makes his predictions for what's next on networking in 2023.
What's next for Cloud and Service Provider infrastructure in 2023?
- Juniper Networks' Sally Bament sees sustainability becoming a key factor shaping carrier network infrastructure.
- She examines what is happening in metro access and aggregation.
Full summary
Has sustainability really become a key factor shaping the market for carrier network infrastructure? What is happening in metro access and aggregation? Here are 3 predictions from Juniper Networks's Sally Bament.
What's next at OCP in 2023?
- Open Compute Project's Cliff Grossner spotlights immersion cooling, optical switching, and disaggregated systems.
- He weighs which of these new OCP technologies grab the limelight in 2023.
Full summary
From immersion cooling, to optical switching, and disaggregated systems there are a lot of new technologies arriving at the Open Compute Project Project. Which ones will be in the limelight in 2023? Open Compute Project's Cliff Grossner makes his predictions for OCP in 2023.
Maximize the Efficiency of Private 5G
- Napatech's Charlie Ashton highlights financial challenges scaling private wireless, needing support optimization and lower energy use.
- Offloading the data plane to a SmartNIC boosts packet core performance and reduces data-center costs and energy.
Full summary
Charlie Ashton, Senior Director of Business Development at Napatech, highlighted the financial challenges operators face when scaling up private wireless networks, emphasizing the need for user support optimization and energy consumption reduction. Napatech offers solutions that enhance packet core performance, support more subscribers, and reduce data center costs and energy consumption by offloading the data plane to a SmartNIC, partnering with packet core software vendors and server OEMs to ensure end users can utilize their technology.
Enterprise Networking Alphabet Soup: Mastering Hybrid Workloads in the AI Era
- Cyxtera's Holland Barry notes growing hybrid workloads and rising AI use in data center co-location.
- He emphasizes venue selection, high power density, and advanced cooling for high-performance compute.
Full summary
Holland Barry, Field CTO for Cyxtera, discussed the growing trend of hybrid workloads running in various venues and the increasing use of artificial intelligence in data center co-location spaces. Barry emphasized the importance of venue selection for effective workload management, the high power density requirements, and the need for advanced cooling solutions for high-performance compute workloads.
Prospects for Telco Cloud and Edge
- Kaloom's Hitendra Sonny Soni discusses telco cloud and edge trends including top players, sustainability, and the killer app.
- Topics span 5G workloads, metaverse, AR/VR, OCP, and P4 programmability.
Full summary
Join Jim Carroll of Converge Digest and Kaloom's SVP Marketing and BD, Hitendra Sonny Soni, on a 1-mile stroll in San Jose, California as they discuss trends in Telco Cloud and Edge including: top players, sustainability, killer app, early advantage, 5G workloads, metaverse, AR/VR, OCP, P4 and more.
CBRS Powers Private Networks
- Federated Wireless CEO Iyad Tarazi points to renewed momentum for private networks at MWC Americas 2022.
- He highlights the role of CBRS in powering those private networks.
Full summary
There is renewed energy at MWC Americas 2022 in Las Vegas – especially for private networks. Iyad Tarazi, president and CEO of Federated Wireless, talks about the momentum and the role of CBRS.
Open RAN accelerates the pace of innovation
- Fujitsu's Rob Hughes says Open RAN accelerates innovation, improving energy efficiency and optical transport.
- He discusses new network elements emerging from the shift.
Full summary
Open RAN is accelerating the pace of innovation, leading to better energy efficiency, improved optical transport, and new network elements. Rob Hughes, Head of Wireless Marketing, Fujitsu Network Communications, discusses the latest advancements.
Rethinking Data Center Fabrics
- Enfabrica's founders argue data center fabric architecture must be rethought to overcome I/O scaling limits.
- Their approach maps to OCP Summit themes of openness, efficiency, scalability, and sustainability.
Full summary
Roshan Sankar and Shrijeet Mukherjee, founders of Enfabrica, a Silicon Valley stealth start-up, discuss how we need to rethink fabric architecture to overcome challenges in I/O scaling and address the key themes at #OCPSummit2022 of openness, efficiency, scalability, and sustainability.
Massive power efficiency gains for tape storage vs HDDs
- IBM's Diamondback Tape Library delivers major power-efficiency gains over HDDs for archival storage.
- The air-gapped, high-density library holds hundreds of petabytes and guards against ransomware.
Full summary
Shawn Brume discusses the power efficiency advantage of IBM’s Diamondback Tape Library over HDDs. Diamondback is a high-density archival storage solution with hundreds of petabytes of capacity and is physically air-gapped to protect against ransomware and other cyber threats.
Smarter chassis designs for sustainable data centers
- Inspur's Alan Chang details sustainable chassis designs using renewable energy, recycling, and thermal reuse.
- Liquid-cooling technologies help reduce data center water consumption.
Full summary
Alan Chang, VP, Technical Operations, Inspur, discusses their showcase at #OCPSummit2022 with a focus on data center sustainability, including utilizing renewable energy, recycling, thermal reuse, and liquid-cooling technologies to reduce water consumption.
OCP adds Sustainability to its Charter
- OCP Chair Rebecca Weekly wraps up the 2022 San Jose Summit and its high-energy atmosphere.
- OCP added Sustainability to its charter this year.
Full summary
“You can feel the energy!” Rebecca Weekly, Chair of the Open Computer Project, provides a wrap-up of this year’s #OCPSummit20222 in San Jose, California, including the addition of Sustainability to its charter.
OCP's Future Technology Symposium
- OCP's Lesya Dymyd shares highlights from the Future Technology Symposium spanning academia, startups, and industry.
- Submitted papers covered five tracks: Network, HPC, Data Center Efficiency, Systems, and Power and Sustainability.
Full summary
OCP’s Lesya Dymyd shares highlights from the Future Technology Symposium at the OCP Global Summit 2022 which invited authors from academia, startup, industry and venture communities to submit papers across five technological tracks: Network, High Performance Computing (HPC), Data Center Efficiency, Systems (server, storage, switches) and Power and Other Data Center Sustainable Solutions.
Advancing Data Center Sustainability
- Schneider Electric's Alexander Rakov describes OCP community efforts to help data centers grow sustainably.
- He sees the community driving innovation toward large gains in sustainable performance.
Full summary
It is imperative that data centers grow sustainably and the Open Computer Project is working to advance this cause. Alexander Rakov, Sustainability Leader – C&SP, Schneider Electric, talks about how the OCP community is driving innovation that could achieve huge gains in sustainable performance.
Energy-Efficient Data Centers, Healthcare, Mobility
- DuPont's Jake Joo expects energy-efficient data centers, healthcare, and mobility to drive 2022.
Full summary
Jake Joo, Technical Manager, DuPont Silicon Valley Technology Center, shares predictions for 2022. Download AvidThink's 2021 Select Report Collection to explore strategies and opportunities in the Telco, Private Mobile Networks, Service Assurance, SD-WAN and Open RAN space.
Edge Computing, Composable, Silicon, and Data Center Sustainability
- Open Compute Project's Cliff Grossner highlights edge computing, composability, silicon, and data center sustainability for 2022.
Full summary
Cliff Grossner, Vice President, Market Intelligence at Open Compute Project, shares predictions for 2022. Download AvidThink's 2021 Select Report Collection to explore strategies and opportunities in the Telco, Private Mobile Networks, Service Assurance, SD-WAN and Open RAN space.