Reports
AI-generated structured vendor updates
WhiteFiber and DriveNets Achieve 111.2 Tbps Cross-DC AI Fabric, Breaking Power Constraints
WhiteFiber announces Project Redwood, partnering with DriveNets Ethernet AI fabric (FSE, VOQ, deep buffers), WEKA storage, and NVIDIA H200 GPUs, achieving 111.2 Tbps bandwidth and 0.9ms latency over 83km dark fiber, treating two geographically separated GPU clusters as a single logical supercluster. Commercialization planned for Q3 2026.
Huawei Ascend 10K-Card Cluster Goes Live, UnifiedBus Protocol Pools All Resources
Huawei launched an Ascend 10,000-card AI cluster in Shaoguan, Guangdong, and showcased the Atlas 950 SuperPoD with its proprietary UnifiedBus interconnect supporting 8,192 NPUs at 16.3 PB/s. Huawei Cloud also entered the Gartner 2026 Cloud AI Infrastructure Leaders quadrant, reinforcing its push for a self-contained AI ecosystem.
NVIDIA Kyber NVL144 Delayed to 2028: Midplane PCB Manufacturing Becomes AI Scaling Bottleneck
SemiAnalysis reveals NVIDIA's Kyber NVL144 delayed beyond 12 months to 2028 due to 78-layer Orthogonal Backplane manufacturing challenges. The interim NVL72x2 solution is cancelled due to operational burdens, and the 4-die Rubin Ultra is also scrapped, leaving a product gap in NVIDIA's scaling roadmap.
NVIDIA AI Compute Partnership: Revenue Share and Credit Backstop to Lock Cloud Providers into DSX AI Factories
NVIDIA launches AI Compute Partnership with revenue sharing and credit backstop, shifting from hardware sales to recurring service revenue. Initial projects include 40K GB300 chips for Sharon AI and 170K GPUs for Firmus, totaling 200K+ high-end chips. NVIDIA is becoming the 'central bank' of AI compute, squeezing cloud brokers.
NVIDIA Rubin Mandates 100% Liquid Cooling with 45°C Warm Water, Reshaping Data Center Thermal Design
NVIDIA reveals Rubin platform's full liquid cooling design: 100% liquid, 45°C warm water inlet, eliminating chillers and fans. Mass production starts H2 2026, with a mandate for all data centers to transition to liquid cooling, marking a definitive shift in AI thermal management.
Huawei and Hubei Mobile Validate AI Inference Acceleration: External KV Cache Boosts Throughput 372%
Huawei and Hubei Mobile completed the first operator AI inference acceleration trial, using OceanStor A800 storage and Ascend A3 supernode with UCM to externalize KV Cache to PB-level storage, achieving up to 372% TPS improvement for long-context inference on GLM-5.1 and MiniMax M2.5 models.
Google Cloud and Nokia Embed Gemini AI Agents to Seize Network Operations Control Plane
Google Cloud and Nokia partner to embed Gemini AI agents (including Router Agent, Event Triage Agent) into Nokia Assurance Center, launching as SaaS on Google Cloud Marketplace in September 2026. Aiming to reduce troubleshooting time by 50-80%, this marks a fundamental shift from rule-based to AI-driven telco operations.
Ericsson Abandons Full-Stack for Modular DMP, IBM watsonx Orchestrate Enters Telco Automation
Ericsson modularizes its Digital Monetization Platform (DMP), cedes CRM front-end to Salesforce, and brings in IBM as system integrator with watsonx Orchestrate multi-agent orchestrator, retreating to software licensing to address 5G ROI pressure.
Ericsson Q1 Earnings: AI-native Radio Reshapes RAN Architecture
Ericsson Q1 earnings show EBITA margin doubling to 21%, alongside the launch of AI-native Radio architecture, transforming 5G base stations from transport pipes to edge inference platforms.
Microsoft Partners with Domestic Operators to Build Sovereign AI Infrastructure in Japan
Microsoft announced a $10B investment in Japan over four years, with a key pillar being a collaboration with Sakura Internet and SoftBank. This partnership will offer GPU-based AI compute services through Azure, managed by domestic providers to ensure data residency within Japan. This addresses the demand for sovereign AI infrastructure for sensitive workloads.
Amazon Advocates EPFD Spectrum Rule Modernization to Unleash Satellite Capacity
Amazon's Leo satellite project demonstrated speeds exceeding 1Gbps, highlighting that current EPFD rules based on 1990s assumptions restrict Ka-band efficiency. The company is advocating rule updates to eliminate unnecessary avoidance zones, improving spectrum efficiency by 180% and reducing operational costs.
NVIDIA Unveils Physical AI Data Factory Blueprint and Frontier Models
NVIDIA launched three physical AI frontier models and an open Physical AI Data Factory reference architecture at GTC 2026, converting computation into synthetic training data via Cosmos world model and OSMO operators. The Omniverse DSX digital twin blueprint enables validation and real-time AI inference integration with Jetson modules.
Ericsson Signs Core Network Modernization Contract with SoftBank in Japan
Ericsson signs a core network expansion and modernization agreement with SoftBank Japan, utilizing dual-mode 5G core solutions to enhance network capacity and reliability. Ericsson provides cloud-native core products including user data management and policy control, enabling SoftBank's transition to 5G SA and network slicing.
Nokia Deploys First AWS Cloud-Native Core Network in ASEAN
Nokia deployed ASEAN's first cloud-native core network on AWS for Tune Talk in Malaysia, using its cloud-native core software to enable rapid service deployment and capacity scaling, accelerating network modernization.
Ericsson and SK Telecom Partner on AI-RAN and 6G Network Innovation
Ericsson and SK Telecom signed an MoU to jointly develop AI-RAN, network slicing, and cloud-native architectures. They will establish a joint innovation center in Korea to explore AI-driven network optimization and automation. The collaboration aims to build foundational technologies for 6G networks.
Nokia Demonstrates AI Automation and Quantum-Safe Networking Direction
Nokia demonstrated AI-driven network automation capabilities for autonomous optimization and fault prediction at OFC 2024. The company also introduced quantum-safe networking solutions to protect fiber infrastructure from future quantum computing threats, showcasing its dual focus on intelligent operations and foundational security.
Bridged Broadband Deploys Nokia 800G Backbone for Rural Connectivity
Bridged Broadband deploys Nokia's 800G IP and optical transport solution to build a 2,500-mile regional backbone with 47 PoPs. Using a consortium model to aggregate multiple provider infrastructures, it delivers carrier-grade broadband to rural areas. The network supports edge data center AI connectivity and high-bandwidth applications, managed through Nokia NSP and WaveSuite.
Ericsson deploys private 5G for airports via VNPT partnership
Ericsson partners with VNPT to deploy private 5G network architecture for airport operations, providing high-bandwidth low-latency connectivity. The solution targets critical scenarios like baggage handling and ground services, establishing a scalable network foundation.
Ericsson Integrates AI into Physical Layer Receiver Algorithms for Spectrum Efficiency
Ericsson launched AI-powered receiver technology that uses AI models to estimate and compensate for wireless channel impairments in real-time, improving spectrum efficiency by up to 50% in harsh conditions. Integrated into Ericsson's radio system portfolio, deployment is planned for 2026.
Ericsson Predicts AI Devices Will Drive Uplink-Dominant Network Traffic
Ericsson forecasts AI devices will increase uplink traffic share from 20% to 60% by 2026, driven by continuous sensor data and video streaming uploads. This necessitates network architecture reassessment and uplink enhancement technologies like carrier aggregation and 6GHz spectrum.