/
Navigation
Chronicles
Browse all articles
Explore
Semantic exploration
Research
Entity momentum
Nexus
Correlations & relationships
Story Arc
Topic evolution
Drift Map
Semantic trajectory animation
Posts
Analysis & commentary
Pulse API
Tech news intelligence API
Browse
Entities
Companies, people, products, technologies
Domains
Browse by publication source
Handles
Browse by social media handle
Detection
Concept Search
Semantic similarity search
High Impact Stories
Top coverage by position
Sentiment Analysis
Positive/negative coverage
Anomaly Detection
Unusual coverage patterns
Analysis
Rivalry Report
Compare two entities head-to-head
Semantic Pivots
Narrative discontinuities
Crisis Response
Event recovery patterns
Connected
Search: /
Command: ⌘K
Embeddings: large
TEXXR

Chronicles

The story behind the story

days · browse · Enter similar · o open

Broadcom unveils Jericho4, a new version of its networking chip that can connect 1M+ processors across multiple data centers and handle distances of over 100km

Dina Bass / Bloomberg :

Bloomberg Dina Bass

Context & Ripple Effects

Broadcom had already moved its data-center networking lineup forward with Tomahawk 6 switch-chip shipments aimed at improving GPU utilization. Jericho4 extends that arc from switching within a data center toward linking processor fleets spread across sites.

The announcement also sits in a widening contest over long-distance AI-network plumbing: Cisco later positioned its Silicon One P200 and 8223 routing systems around faster transfers over long optical cables.

First-order effects

  • Jericho4 gives Broadcom a networking-chip offering designed for deployments that need to connect more than one million processors across multiple data centers separated by more than 100 km.
  • Data-center operators and system builders evaluating geographically distributed processor clusters gain another Broadcom option for the inter-data-center network layer.

Second-order effects

  • The product raises competitive pressure on routing and switching suppliers to demonstrate comparable scale and long-distance optical connectivity; Cisco’s subsequent Silicon One launch underscores that contest.
  • Network design choices become more consequential for customers spreading compute across sites, because the fabric connecting those sites can affect how effectively processors operate as one pool.

Third-order effects

  • If these offerings gain adoption, AI infrastructure may increasingly be designed as geographically distributed compute fabrics rather than isolated data centers, elevating networking silicon alongside processors as a strategic bottleneck.
  • The market could favor suppliers able to pair high-scale switching with long-haul connectivity, reinforcing an integrated infrastructure stack rather than point products.

The trend: AI infrastructure is expanding from dense in-data-center clusters toward network fabrics that can coordinate compute across campuses and distant sites.