Broadcom
Maker of the Tomahawk and Jericho switch chips used in most data center networks, and owner of VMware.
Recent activity
Stories per weekLatest stories
10 most recent-
Broadcom pitches disaggregated Ethernet fabric against Cisco Silicon One
Broadcom unveiled its open Ethernet fabric portfolio at SC26, contrasting its disaggregated ASIC strategy (Tomahawk Ultra, Tomahawk 6, Jericho 4 paired with Thor Ultra NICs) against Cisco's unified P4-programmable Silicon One architecture for AI workloads.
Why it matters Clarifies architectural divergence in 102.4T switching: disaggregated ASICs versus unified silicon affects tail-latency behavior, fabric programmability, and multi-vendor interoperability for large AI clusters.
-
Broadcom pushes Scale-Up Ethernet Transport for multi-vendor AI fabrics
Broadcom hosted SC26 workshop coverage of open AI and HPC fabric standards, emphasizing Scale-Up Ethernet Transport (SUE-T) under Open Compute Project and Ultra Ethernet Consortium specifications for deterministic, lossless transport, load balancing, congestion control, resilience, and telemetry across multi-vendor environments.
Why it matters SUE-T standardization under Open Compute Project and UEC 1.0 certification enable practitioners to architect multi-vendor AI fabrics with deterministic transport guarantees and certified interoperability; vendors can now ship UEC-compliant fabric switches, NICs, and telemetry stacks without proprietary lock-in.
-
Broadcom pitches disaggregated AI Ethernet fabric against Cisco Silicon One
Broadcom presented its disaggregated AI fabric strategy at SC26 combining Tomahawk 6 (102.4 Tbps), Tomahawk Ultra (low-latency intra-rack), Jericho 4 (deep-buffer routing), and Thor Ultra 800G NIC with UEC compliance, contrasting with Cisco's unified Silicon One approach.
Why it matters Network architects must choose between Broadcom's specialized role-based ASICs (optimized per function) versus Cisco's unified programmable silicon; Broadcom's approach reduces switching latency and power per function, critical for 100k+ GPU clusters.
-
Broadcom shows Tomahawk 6 and Thor Ultra at SC26
Broadcom demonstrated Tomahawk 6 and Thor Ultra 800G Ethernet silicon at SC26, highlighting compatibility with Ultra Ethernet Consortium (UEC) specifications for AI workloads. Thor Ultra eRoCE NICs achieve job completion times matching 800G InfiniBand quantum fabrics by placing packets directly into GPU memory.
Why it matters Open Ethernet silicon with UEC compliance and direct GPU memory placement eliminates packet ordering dependencies, allowing switches to spray packets across all links and closing the performance gap with InfiniBand at Ethernet scale and cost.
-
Broadcom uses topology-aware correlation to untangle post-merger networks
Broadcom published a case study demonstrating how topology-aware root cause analysis and multi-vendor event correlation filters extraneous noise to isolate true origins of network degradation and pinpoint device and link dependencies during post-acquisition network integration.
Why it matters Topology-aware root cause analysis reduces MTTR by automatically correlating multi-vendor device events, suppressing extraneous symptom alarms, and pinpointing exact failure locations across hundreds of thousands of endpoints.
-
CISA flags VMware ESX and Linux kernel flaws as exploited
CISA added two new vulnerabilities to its Known Exploited Vulnerabilities Catalog on October 28, 2026, based on evidence of active exploitation: CVE-2026-41709 (Broadcom VMware ESX Insufficient Logging) and CVE-2026-46300 (Linux Kernel Missing Initialization of a Variable).
Why it matters Both CVEs are in CISA KEV as actively exploited; infrastructure teams running VMware hypervisors and Linux systems must prioritize patching to prevent adversary activity from going undetected and to prevent memory corruption exploitation across production environments.
-
Meta open sources liquid-cooled 102.4T AI network racks
Meta shared networking innovations supporting its latest AI models and gigawatt-scale Prometheus cluster at OCP Summit 2026. IcePack and IceCube are Meta's liquid-cooled network racks featuring 102.4T switch blades with 128×800G ports using Broadcom Tomahawk 6C and NVIDIA Spectrum-6 variants, running FBOSS.
Why it matters Meta's liquid-cooled Tomahawk 6C and Spectrum-6 fabric platforms at 102.4T capacity demonstrate production-ready 800G deployment patterns for hyperscale AI clusters at gigawatt scale.
-
Ultra Ethernet aims to match InfiniBand as Broadcom silicon nears
The Ultra Ethernet Consortium, formed in 2023 with AMD, Intel, HPE, Broadcom, Cisco, Meta, and Microsoft as founding members, aims for an open Ethernet RDMA fabric that matches InfiniBand latency without proprietary components. UEC 1.0 (published 2024) defines credit-based reliability in NIC firmware, eliminating head-of-line blocking. Broadcom's Tomahawk-UEC silicon is expected in 2026–2027.
Why it matters UEC 1.0 eliminates head-of-line blocking that makes RoCEv2 hard to operate at scale. This October 19 update provides operators a current decision framework: UEC hardware is still pre-production, making Spectrum-X the best production Ethernet option today if you need performance beyond vanilla RoCE.
-
KDDI and Broadcom test co-packaged optics in Japan
KDDI and Broadcom announced a collaboration testing co-packaged optics (CPO) technology at a KDDI data center using Broadcom's Tomahawk 6 Ethernet switch chip with Davisson optical engines. The testing focuses on GPU-to-GPU communications for AI data centers.
Why it matters CPO transitions from research prototype to production validation by tier-1 telecom operators, accelerating adoption timelines for power-efficient, high-density AI fabrics in commercial deployments outside hyperscaler-only environments.
-
IP Infusion OcNOS adds Broadcom Tomahawk 6 at 102.4 terabits
IP Infusion announced OcNOS-DC support for Broadcom Tomahawk 6, enabling 102.4 Tbps fabric with 512 ports of 200G, supporting scale-up clusters of 128,000 XPUs in a two-layer leaf-spine topology.
Why it matters Eliminates super-spine layers and reduces hop count, cutting latency and optics cost while maintaining lossless RoCEv2 at full line rate for AI workloads.
Coverage history
32 earlier storiesOctober 2026 17
- Oct 17IP Infusion's OcNOS adds support for Broadcom Tomahawk 6Business Wire
- Oct 15Broadcom demos Tomahawk 6 co-packaged optics at OCP SummitIEEE Comsoc
- Oct 15Celestica ships 102.4T liquid-cooled switch with co-packaged opticsStorageNewsletter
- Oct 14Micas launches three 102.4T Tomahawk 6P Ethernet switchesGlobe Newswire
- Oct 14Fuji automates SMT placement of Broadcom co-packaged optics switchElectronics Media
- Oct 13Micas plans Tomahawk 6 white-box switches for Q4StorageReview
- Oct 12Micas launches three 102.4T Tomahawk 6 Ethernet switchesGlobeNewswire
- Oct 10Broadcom guide urges consolidating network observability tools for AIBroadcom
- Oct 10Broadcom to show Tomahawk 6 and co-packaged optics at OCPAIwire (HPC Wire)
- Oct 10Arista scales Ethernet fabric to 144 accelerators per domainArista Networks
- Oct 10Broadcom pitches scale-up and scale-across AI networking at OCPGlobe Newswire
- Oct 8Broadcom pushes scale-up, scale-out and scale-across AI networkingBroadcom press release
- Oct 8Arista launches 1.6T rack-scale AI switches with liquid coolingStreet Insider
- Oct 8Broadcom starts volume shipments of Davisson co-packaged opticsData Centre & Network News
- Oct 3HPE lifts AI networking order target to $3 billionYahoo Finance
- Oct 2102.4T switch silicon ships as 1.6T Ethernet nearsSemiconductor Insight
- Oct 1Broadcom Tomahawk 6 reaches volume production at 102.4 TbpsIPinfusion
September 2026 2
- Sep 29Multi-vendor 1.6T Ethernet interoperability validated at OFC 2026Hengtong
- Sep 3Broadcom Launches VMware AI Factory: Bare-Metal to Model in Hours via MetalSoft IntegrationBroadcom / VMware Newsroom
August 2026 5
- Aug 20AI Data Center Networking: Scale Up, Out, and Across with 102.4T EthernetData Center Knowledge
- Aug 16Arista Networks Q2 2026: $3.04B Revenue on AI Networking Acceleration, Q3 Guidance Raised to $3.30BYahoo Finance / Arista Networks
- Aug 11Anthropic Locks In $71B in Compute Commitments Across AWS, Google, Broadcom Amid Revenue AccelerationAnthropic Blog
- Aug 42026 Data Center Networking Report: NVLink Scales Rack-Wide, Ethernet Reaches 100K+ Clusters, Standards ConsolidateNext Gen Infrastructure (NGI) Research Brief
- Aug 2AMD Advancing AI 2026: Helios Rack Pricing at $5.25M, UALink Adoption as Critical Swing FactorTech Insider
July 2026 6
- Jul 202026 Data Center Networking – NextGenInfra InsightsNextGenInfra
- Jul 15Redefining Network Operations in the Age of Agentic AI: Maintaining Engineer Expertise During Autonomous TransitionsBroadcom Network Observability
- Jul 13Broadcom showcases industry-leading AI infrastructure solutions at OFC 2026Broadcom Inc.
- Jul 9What's New in Network Observability for Summer 2026: Agentic AI, Alert Noise Reduction, Device Lifecycle AutomationBroadcom Academy Blog
- Jul 9GigaOm 2026 Radar: Network Observability Market Shift to Unified Intelligence and Automated Noise ReductionBroadcom Network Observability
- Jul 2When the GPUs are waiting for the networkAixia