GammaCORE UCIe Die-to-Die Controller IP
GammaCORE is a highly configurable and customizable Universal Chiplet Interconnect Express (UCle™) Die-to-Die Controller IP implementing the latest UCIe 2.0 specification and supporting UCIe Streaming protocol applications.
With Alphawave’s AresCORE D2D PHY IP, GammaCORE provides the complete UCIe solution for an open and robust chiplet ecosystem.
GammaCORE UCIe Die-to-Die (D2D) Controller IP
GammaCORE UCIe D2D Controller IP enables multi-die applications, such as the connection of I/O chiplets to a main die, accelerator die to a processor die, multi-die packet processors, multi-die Ethernet switches, etc.
GammaCORE Controller IP core consists of the Streaming Protocol Layer, to extend the SoC interface across the UCIe link, and the Adapter Layer, to provide a reliable end-to-end link.
On the UCIe link side, GammaCORE provides seamless connection to the AresCORE UCIe D2D PHY IP through the RDI interface. On the system side, GammaCORE connects to one or multiple internal SoC interfaces.
High Configurability and Customizability
GammaCORE supports
- Any specific SOC Interface: AXI4, AXI-S, TileLink, CXS, CHI, and some proprietary interfaces over UCIe.
- Flit and Raw mode operation.
- Configurable datapath width to address specific Bandwidth and target clock frequency requirements.
- Combination with other Alphawave Semi companion IP cores to enable many different configurations.
Comprehensive Verification
The GammaCORE Controller IP has been verified against leading UCIe Verification IPs (VIP). The IP is silicon-proven, together with Alphawave’s AresCORE UCIe D2D PHY, on variety of test chips in leading-edge foundries across multiple technology nodes.
Typical Applications
The AresCORE UCIe D2D interconnect enables SoC developers to break the boundaries of the reticle limit by including multiple dies in the same package. This multi-die approach to silicon design allows for better yielding, smaller purposeful dies vs. traditional SoCs. The main target applications AresCORE UCIe D2D interconnect include, but are not limited to:
- AI accelerators
- Server class CPUs
- Network switches designed for large compute
- FPGAs
- 5G base stations
- IO and optical transceivers
Additional Resources
- Press Releases:
- Alphawave Semi Scales UCIe™ to 64 Gbps Enabling >20 Tbps/mm Bandwidth Density for Die-to-Die Chiplet Connectivity
- Unveiling the Industry’s First Silicon-Proven 3nm, 24Gbps UCIe™ IP Subsystem with TSMC CoWoS® Technology
- Alphawave Semi Drives Innovation in Hyperscale AI Accelerators with Advanced I/O Chiplet for Rebellions Inc.
- Alphawave Semi Tapes Out Industry-First, Multi-Protocol I/O Connectivity Chiplet for High-Performance Compute and AI Infrastructure
- Alphawave Semi Collaborates with Arm on High-Performance Compute Chiplet
- Alphawave Semi Demonstrates 3nm Silicon-Proven 24Gbps Universal Chiplet Express™ (UCIe™) Subsystem for High-Performance AI Infrastructure
- Alphawave Semi Spearheads Chiplet-Based Custom Silicon for Generative AI and Data Center Workloads with Successful 3nm Tapeouts of HBM3 and UCIe IP
- Alphawave Semi Elevates Chiplet-Powered Silicon Platforms for AI Compute through Arm Total Design
- On-Demand Videos: