Dual Port Register File Compiler (1 Read-Only Port, 1 Write-Only Port)
Movellus Debuts Industry-First On-Die Power Delivery Network Analyzer
Aeonic Insight PDN IQ arms industry with unprecedented transistor-level PDN visibility and analytics that spans the entire product cycle from bench to in-field monitoring
SANTA CLARA, Calif., May 01, 2025 — Movellus today announced the industry’s first on-chip sensor for power delivery network (PDN) characterization and monitoring. Power integrity is vital to the performance and efficiency of any system-on-chip (SoC). However, the industry has very limited visibility into the behavior of on-chip PDNs, which can result in voltage over-margining (compromising both power and performance), missed performance targets, and higher DPPM rates. And as the industry evolves to larger SoCs, stacked die, and chiplets, the problem is further magnified, making PDN visibility even more critical to overall system performance and energy efficiency. PDN IQ enables improvements in power, performance, and uptime for advanced semiconductors by providing PDN visibility across the entire silicon lifecycle from bench characterization and production test to in-field monitoring.
“Robust PDN operation is required to ensure the functionality and performance of high end SoCs. Today, the industry relies on a variety of black-box techniques to evaluate these critical networks, but these approaches are becoming less effective in the face of ever larger devices, advanced packaging, and severe power swings in workloads.” said Ranganathan “Suds” Sudhakar, Chief Development Officer at d-Matrix. “Having direct visibility of PDN efficacy, anywhere on chip, is a compelling mechanism to optimize the power-performance characteristics of large SoCs and will be a game changer for the industry.”
“The insatiable demand for greater bandwidth in AI accelerators continues to expand the quantity of processor and memory chiplets integrated within a single device. As disparate semiconductors are knitted together with advanced packaging technologies, such as Deca’s M-Series fan-out, optimized power delivery is an inherent challenge.” commented Tim Olson, CEO of Deca Technologies, “The introduction of this exciting new in-situ power delivery network telemetry from Movellus gives customers a breakthrough capability to optimize voltage-frequency points to maximize performance while improving system quality & reliability.”
Aeonic Insight PDN IQ delivers high-speed, transistor-level voltage telemetry combined with sophisticated trigger and trace capabilities,
and advanced statistics that deliver value throughout the entire silicon lifecycle. Some of the use cases include:
- Refining voltage-frequency curves during characterization
- Observing on-die PDN behavior under high stress workloads on the bench & in field
- Improve screening accuracy to reduce DPPM rates in production
- Improving fleet uptime by continuously monitoring PDN behavior and flagging anomalies
“SoCs are growing exponentially in complexity, especially for the datacenter and AI segments. Design teams need to turn over every rock in their designs to optimize for power and performance, and this includes the power delivery network,” said Rich Wawrzyniak, Principal Analyst at SHD Group. “Movellus is arming its customers with the visibility and telemetry required to optimize SoC power with their new on-die PDN analyzer IP.”
Aeonic Insight™ PDN IQ can also integrate with third-party analytics platforms, offering even greater flexibility. Silicon teams can feed telemetry data from the trace buffer and other registers available in the IP into an analytics stack to gain deeper insights into PDN behavior, further enhancing design and deployment strategies.
Movellus continues to advance high-performance silicon through feature-rich IP. The Aeonic™ Digital IP platform has been integrated by multiple customers across various process nodes, with applications ranging from performance-centric cloud datacenter compute and AI offerings to ultra-low power edge AI devices. This latest milestone enables advanced on-die observability for power delivery networks, ensuring maximum power efficiency for cutting-edge SoCs, AI accelerators, and processors. Learn more about this latest milestone at www.movellus.com/aeonic-insight-pdn-iq.
About Movellus
Movellus provides critical technology that is integrated into an array of applications ranging from edge AI devices to performance-centric cloud datacenter compute and networking offerings. The company is headquartered in Santa Clara, CA, with R&D centers in Michigan and Toronto. Visit us at: www.movellus.com
|
Related News
- Movellus Introduces Aeonic Power™ Product Family for On-Die Voltage Regulation
- Movellus Introduces Aeonic Insight™ Product Line for On-die Telemetry
- Movellus Announces Industry-First Integrated Droop Response System for SoCs
- Xilinx Debuts Industry-First P4_16 to FPGA Compilation and Introduces New P4-NetFPGA Workflow for Networking Researchers at P4 Industry Events
- Xilinx Debuts Industry-First Solutions at OFC 2017 and Further Expands High Speed Data Center Interconnect Offerings
Breaking News
- Weebit Nano Q3 FY25 Quarterly Activities Report
- Codasip launches complete exploration platform to accelerate CHERI adoption
- VSORA Raises $46 Million to Bring World's Most Powerful AI Inference Chip to Market
- Arteris Joins Intel Foundry Accelerator Ecosystem Alliance Program to Support Advanced Semiconductor Designs
- SkyeChip Joins Intel Foundry Accelerator IP Alliance
Most Popular
- Arteris Joins Intel Foundry Accelerator Ecosystem Alliance Program to Support Advanced Semiconductor Designs
- SkyeChip Joins Intel Foundry Accelerator IP Alliance
- CFX 0.13μm eFuse OTP IP has been applied in the mass production of over 15,000 CMOS image sensors
- Siemens and Intel Foundry advance their collaboration to enable cutting-edge integrated circuits and advanced packaging solutions for 2D and 3D IC
- Cadence Expands Design IP Portfolio Optimized for Intel 18A and Intel 18A-P Technologies, Advancing AI, HPC and Mobility Applications
![]() |
E-mail This Article | ![]() |
![]() |
Printer-Friendly Page |