World's First Standards-Compliant 112G PHY IP for Linear Optics: A Turning Point for AI Interconnects
Synopsys delivers standards-compliant 112G PHY IP for linear optics, helping reduce AI interconnect power, latency, and heat.
Vendor:
Synopsys, Inc.
Category:
Custom
Die-to-Die, 112G Ultra-Extra Short Reach PHY in GF (12nm)
The Synopsys XSR PHY IP for 112Gbps per lane die-to-die connectivity enables high-bandwidth ultra and extra short reach interface…
Overview
The Synopsys XSR PHY IP for 112Gbps per lane die-to-die connectivity enables high-bandwidth ultra and extra short reach interfaces in multi-chip modules (MCMs) for hyperscale data center, AI, and networking applications. The lowlatency, low-power, and compact PHY supports NRZ and PAM-4 signaling from 2.5G to 112G data rates and is compliant with the OIF CEI-112G and CEI-56G standards for extra- short reach (XSR) links. The XSR PHY offers flexible layout for maximum bandwidth per die-edge by allowing placement of the square macros along all edges of the die. It deploys 16-lane transmit and receive macros for optimized segmentation on the multiple dies. The robust DLL-based clock forwarded architecture enables high energy efficiency while supporting reliable links of up to 50 millimeters for large MCMs. The PHY enables multi-die connectivity over organic substrates, which helps reduce packaging costs without requiring advanced interposer-based packaging over shorter distances. The embedded bit error rate (BER) tester and nondestructive 2D eye monitor capability provide on-chip testability and visibility into channel performance. Besides the PMA and PMD, the PHY includes a raw-PCS to facilitate the interface with the on-chip network, regardless of the existing networking protocol. The XSR IP is combined with Synopsys’ comprehensive routing feasibility analysis, packages substrate guidelines, signal and power integrity models, and crosstalk analysis for fast and reliable integration into SoCs
Key features
- Offers leading performance, power, and area / beachfront per terabit
- Includes 16 lanes of NRZ and PAM-4 transmitters or receivers
- Targeting the OIF XSR standards: CEI-112G and CEI-56G
- Implements robust clock forwarded and embedded clock recovery algorithms for additional flexibility
- Physical Medium Dependent (PMD) soft-layer includes power sequencing and calibration engines
- Implements RX scope and PRBS and pattern generator, matcher and error counter for BERT testing
- Non-destructive 2D eye monitor capability can be performed during mission-mode operation
- Shared PLLs allow maximum energy efficiency
- Optimal receiver jitter tolerance supports a wider range of board layout designs, immunity to interference (cross talk), and reduces design constraints on board signal path
Benefits
- 16-lane TX and RX square macros for placement in any edge of the die
- Supports 2.5G to 112G data rates, enabling very high bandwidth per mm of beachfront for die-to-die and die-to-optical engine connectivity
- Implements NRZ and PAM-4 signaling
- Meets the performance, efficiency and reliability requirements of die-to-die interconnects
- Robust DLL-based, clock forwarded architecture minimizes complexity and power dissipation
- Linear equalization and T-Coils in RX and TX allow compliance to XSR links up to 50mm for large MCM designs
- Low jitter phase-locked loops (PLLs) provide robust timing recovery and better jitter performance
- Targeting the OIF CEI-112G and CEI- 56G standards for XSR links
Applications
- Hyperscale data center
- Enterprise and campus networks
- Cloud computing/networking
- Service provider networks
- Artificial intelligence and machine learning
What’s Included?
- Verilog models and test bench
- Protocol-specific test bench
- Liberty timing views (.lib), LEF abstracts (.lef), CDL netlist (.cdl)
- GDSII
- IP-XACT XML files with register details
- ATPG models
- IBIS-AMI models
- Documentation
Silicon Options
| Foundry | Node | Process | Maturity |
|---|---|---|---|
| GlobalFoundries | 12nm | LP | Available on request |
Specifications
Identity
Part Number
dwc_d2d_sr112_phy_gf
Vendor
Synopsys, Inc.
Type
Silicon IP
Files
Note: some files may require an NDA depending on provider policy.
Provider
HQ:
USA
Learn more about Custom IP core
The Integrated Design Challenge: Developing Chip, Software, and System in Unison
“We are in an era of pervasive intelligence,” said Shankar Krishnamoorthy, chief product development officer at Synopsys. Moderating a panel discussion at the inaugural Synopsys Executive Forum, he described the transformative impact of artificial intelligence (AI) and the challenges ahead.
High Bandwidth Memory (HBM) at the AI Crossroads: Customization or Standardization?
As artificial intelligence (AI) reshapes industries and advances technological frontiers, its success hinges on advanced memory capabilities. Leading this transformation is High Bandwidth Memory (HBM), which offers unparalleled speeds and efficiencies.
Synopsys Interconnect IPs Enabling Scalable Compute Clusters
With more than 25 years of experience in delivering field-proven silicon IP solutions, we are thrilled to partner with NVIDIA and the NVIDIA NVLink ecosystem to enable and accelerate the creation of custom AI silicon. This strategic collaboration will leverage Synopsys' expertise in silicon IPs to assist in the development of bespoke AI silicon, forming the foundation for advanced compute clusters aimed at delivering the next generation of transformative AI experiences.
What’s Your Vector? Synopsys Introduces New ARC VPX6 Digital Signal Processor
The new ARC VPX6 processor introduces 1024-bit vector processing to the lineup, extending the existing VPX family, which includes VPX5 (512-bit), VPX3 (256-bit), and VPX2 (128-bit) variants. As such, VPX6 doubles the attainable peak performance while preserving full software compatibility with the other members of the VPX family. By simply upgrading hardware without any recoding, designers can instantly realize performance gains from day one.
40G UCIe IP Advantages for AI Applications
For AI workloads to be processed reliably at a fast rate, the die-to-die interface in multi-die designs must be robust, low latency, and most importantly high bandwidth. This article outlines the need for 40G UCIe IP in AI data center chips leveraging multi-die designs.
Frequently asked questions about Custom Die-to-Die IP cores
What is Die-to-Die, 112G Ultra-Extra Short Reach PHY in GF (12nm)?
Die-to-Die, 112G Ultra-Extra Short Reach PHY in GF (12nm) is a Custom IP core from Synopsys, Inc. listed on Semi IP Hub. It is listed with support for globalfoundries Available on request.
How should engineers evaluate this Custom?
Engineers should review the overview, key features, supported foundries and nodes, maturity, deliverables, and provider information before shortlisting this Custom IP.
Can this semiconductor IP be compared with similar products?
Yes. Buyers can compare this product with similar semiconductor IP cores or IP families based on category, provider, process options, and structured technical specifications.