analyze
design
deploy

Empowering innovation and scale
that Accelerates Intelligent
Computing everywhere

technology

Intelligent software-driven, customizable, System IP
solutions for efficient, yet unprecedented scale for SoCs
and Chiplets. Removes guesswork, reduces risk and cost
of delivering complex high-performance systems.

Performance & Scale

Hyper-efficient transport fabric that can deliver 4TB/s throughput in a single cluster, and scale to multiple PB/s for multi-chiplet AI applications, and support advanced high bandwidth memories.

Extreme Flexibility

The Network on Chip (NoC) is extremely flexible to be customized to any topology. NoC can be extended to custom protocols that allow for greater innovation, in AI, QoS, Debug and more.

Unified Efficiency

Transport is separate from protocol layers, minimizing wires and logic in building a unified NoC that supports coherent, non-coherent, and custom protocols for greatest efficiency with lowest cost and power.

Data-driven design

Extensive analysis and optimization through software platform enables a NoC system that delivers guaranteed performance on target workloads and algorithmically optimized for future workloads.

Derisked System

Software-driven development ensures the NoC is correct by construction, and deadlock-free with traditional and formal validation, substantially reducing risk for system development.

Modular Implementation

The NoC is developed with modular components, improving flexibility and expediting design delivery. It uses physical design aware tiling approaches, to ease implementation, integration and signoff.

Ready for Chiplets

Industry’s first fabric solution designed for multi-cluster and
multi-chiplet designs

Chiplet Ready Network

  • Verifiable protocol level cluster perimeter
  • Performance and congestion isolation

Chiplet Ready Cache Coherency

  • Multi-level cache coherent fabric IP
  • Supports UMA, NUMA or sub-NUMA

Chiplet Ready Design

  • Co-optimization of cache, memory, IO stacks
  • Data-driven D2D analysis and optimization

products

Our products deliver best-in class system IP and
foundational software to help analyze, design,
optimize and deploy customized single or multi-die
solutions for complex, intellgent computing systems.

WeaverPro Software components can be used standalone for architectural exploration and analysis, and to configure, optimize and design systems with WeaveIP. The performance analysis can run on actual workloads, and the software has SystemC and Python APIs to integrate with your development environment.

  • Built-in accurate, high speed simulator
  • Architecture and micro-architecture analysis and guidance
  • Algorithmic optimization and QoS
  • Correct by construction RTL generation
  • Tiling and physically aware implementation planning

Learn more about WeaverPro

WeaveIP is a portfolio of products built on a common transport that is architected for efficiency and allows for creation of unified fabric solutions that are optimized for minimum power and silicon footprint without compromising performance, and deployed faster with clarity on achieving KPI targets.

  • Advanced multi-level cache coherency with CHI and CXL
  • High speed coherent and non-coherent AMBA 5 design
  • Customizable protocol fabric for AI and more
  • Multicast capability for advanced AI acceleration
  • Reliability, Availability, Serviceability and runtime QoS

Learn more about WeaveIP

markets

Our technology is versatile and applicable
across most applications that need efficient
data movement. Our primary focus is on the
following segments:

HyperscaleHyperscale

Datacenter

HyperscaleHyperscale

Infrastructure

HyperscaleHyperscale

AI Acceleration

HyperscaleHyperscale

Automotive

HyperscaleHyperscale

IoT

Large-scale coherent CPU computation in Hyperscale and Enterprise computing. WeaveIP delivers extensive data throughput at low latency with RAS feature, QoS and chiplet-readiness needed for these applications.

Control plane and dataplane applications in wireless and wired networking infrastructure including CDNs, SLAMs and NanoCells. WeaveIP delivers scalability and adaptability and chiplet readiness for many core and thread-rich systems.

AI accelerators that are scaling to high throughput, parallelised data processing. WeaveIP delivers customizable protocols, multicast capability for many-chiplet systems that scale into PetaBytes.

Advanced ADAS, Cockpit and Autonomous systems. WeaveIP enables unified fabric for high-level integration for efficient and performant coherent and non-coherent systems.

Edge servers, Home gateways, networking hubs, complex consumer devices. WeaveIP enables compact heterogeneous compute platforms for highly-efficient applications.

Partners and Ecosystem

Partner with us

We believe in engaging with committed partners to create a strong ecosystem that synergistically drives progress. We would love to collaborate with you.

Press

Blog

Whitepapers