Products
Extreme-Radix Switching for AI Scale-Up and Scale-Out Systems
AI infrastructure is reaching the limits of traditional switch architectures. As GPU clusters continue to scale, conventional crossbars become increasingly difficult to implement due to wiring complexity, physical design challenges, and limited scalability. Next-generation AI systems require a new approach that delivers non-blocking communication while scaling efficiently across thousands of ports and multiple chiplets.
NeuraScale is Baya's non-blocking switch fabric IP designed specifically for AI scale-up and scale-out systems. Built on a distributed array of replicated switching tiles instead of a monolithic crossbar, NeuraScale delivers crossbar-class performance with dramatically improved scalability, physical design efficiency, and faster time to market.
Supporting UALink™, Ultra Ethernet, AMBA, and UCIe-based chiplet connectivity, NeuraScale enables software-defined, modular switch architectures that can scale from hundreds to thousands of ports while maintaining low latency, predictable bandwidth, and implementation flexibility.
NeuraScale delivers non-blocking performance at extreme scale without the limitations of traditional crossbars.
Decades-old interconnect technology
Extreme-radix switch fabric
Larger Scale-Up Domains
Fewer Scale-Out Switches
Switch Ports (256×256)
Deliver crossbar-class, single-hop communication with predictable low latency and guaranteed non-blocking behavior, even as AI systems scale to thousands of ports.
Scale from hundreds to thousands of UALink™, Ultra Ethernet, or AMBA ports through a distributed tile architecture that supports multi-chiplet AI systems without inherent port limitations.
Simplify implementation using a modular architecture built from a single reusable design tiles, reducing wiring complexity and accelerating physical design and time to market.
Support high-throughput multicast, weighted fair bandwidth allocation, deadlock prevention, and efficient flow control to maximize performance across demanding AI workloads.
Accelerate deployment with integrated debug, monitoring, RAS, ECC protection, runtime repair, and performance management capabilities for reliable operation at scale
Discover how NeuraScale delivers non-blocking, extreme-radix switching that scales from hundreds to thousands of ports while simplifying physical implementation. Request a product brief or connect with our team to discuss your architecture design.
Learn how NeuraScale provides a non-blocking, tile-based alternative to traditional crossbars for extreme-radix AI scale-up and scale-out switches. The brief compares NeuraScale with crossbars and classic mesh fabrics, explains its chiplet-based scalability and physical-design approach, outlines its transport, monitoring, and RAS capabilities, and presents reference designs ranging from 50 Tbps to 800 Tbps.