arXiv subjects

Julien Ryckaert

Publications and source records attributed to Julien Ryckaert.

5 recordsLinked to original sources

arXiv2026

Beyond HBM-on-GPU: Thermal Design Envelope for 3D Volumetric DRAM-on-GPU Integration

The scaling of GPUs for AI and HPC workloads is increasingly constrained by the capacity, bandwidth, and thermal limits of both 2.5D HBM-GPU and direct-stacked 3D HBM-on-GPU integration. This work establishes the thermal design envelope for 3D volumetric DRAM-on-GPU integration, in which vertically oriented DRAM dies and interleaved cooling cavities reshape heat flow and memory interfacing above the GPU. Using a package-level thermal model anchored to a consistent HBM-on-GPU baseline and driven by a realistic reticle-scale non-uniform GPU power map, we quantify the key parameters governing thermal feasibility. Stack height is the dominant limiter of peak temperature, while cooling-cavity conductivity shifts the feasible region, and mold insertion and stack orientation further modulate thermal behavior. A distributed memory-controller and network-on-chip tier introduces only a moderate thermal penalty. Although die-level parallelism increases bandwidth, the reduction in simulated training time saturates once execution becomes compute-bound. These results define a bounded co-design space across bandwidth, capacity, and thermal constraints for 3D volumetric DRAM-on-GPU integration.

cs.ET↗

arXiv2025

CMOS 2.0 -- Redefining the Future of Scaling

We propose to revisit the functional scaling paradigm by capitalizing on two recent developments in advanced chip manufacturing, namely 3D wafer bonding and backside processing. This approach leads to the proposal of the CMOS 2.0 platform. The main idea is to shift the CMOS roadmap from geometric scaling to fine-grain heterogeneous 3D stacking of specialized active device layers to achieve the ultimate Power-Performance-Area and Cost gains expected from future technology generations. However, the efficient utilization of such a platform requires devising architectures that can optimally map onto this technology, as well as the EDA infrastructure that supports it. We also discuss reliability concerns and eventual mitigation approaches. This paper provides pointers into the major disruptions we expect in the design of systems in CMOS 2.0 moving forward.

cs.ET↗

arXiv2025

Physical Design Exploration of a Wire-Friendly Domain-Specific Processor for Angstrom-Era Nodes

This paper presents the physical design exploration of a domain-specific processor (DSIP) architecture targeted at machine learning (ML), addressing the challenges of interconnect efficiency in advanced Angstrom-era technologies. The design emphasizes reduced wire length and high core density by utilizing specialized memory structures and SIMD (Single Instruction, Multiple Data) units. Five configurations are synthesized and evaluated using the IMEC A10 nanosheet node PDK. Key physical design metrics are compared across configurations and against VWR2A, a state-of-the-art (SoA) DSIP baseline. Results show that our architecture achieves over 2x lower normalized wire length and more than 3x higher density than the SoA, with low variability in the metrics across all configurations, making it a promising solution for next-generation DSIP designs. These improvements are achieved with minimal manual layout intervention, demonstrating the architecture's intrinsic physical efficiency and potential for low-cost wire-friendly implementation.

cs.AR↗

arXiv2025

Thermal Implications of Non-Uniform Power in BSPDN-Enabled 2.5D/3D Chiplet-based Systems-in-Package using Nanosheet Technology

Advances in nanosheet technologies have significantly increased power densities, exacerbating thermal management challenges in 2.5D/3D chiplet-based Systems-in-Package (SiP). While traditional thermal analyses often employ uniform power maps to simplify computational complexity, this practice neglects localized heating effects, leading to inaccuracies in thermal estimations, especially when comparing power delivery networks (PDN) in 3D integration. This work examines the thermal impact of non-uniform power distributions on SiPs utilizing frontside (FSPDN) and backside (BSPDN) power delivery approaches. Using high-resolution thermal simulations with non-uniform power maps at resolutions down to 5 micrometers, we demonstrate that uniform power assumptions substantially underestimate peak temperatures and fail to reveal critical thermal differences between BSPDN and FSPDN configurations in 3D scenarios. Our results highlight that BSPDN configurations in 3D, although beneficial in simplified uniform scenarios, exhibit pronounced thermal penalties under realistic, localized workloads due to limited lateral heat spreading. These findings emphasize the necessity of adopting fine-grained, workload-aware power maps in early-stage thermal modeling to enable accurate PDN assessment and informed thermal-aware design decisions in advanced nanosheet-based 3D SiP.

cs.ET↗

arXiv2022

Towards sub-30nm Contacted Gate Pitch, Forked Contact and Dynamically-Doped Nanosheets to Enhance Si and 2D Materials Device Scaling

We propose a novel Forked-Contacts, Dynamically-Doped Multigate transistor as ultimate scaling booster for both Si and 2D materials in aggressively-scaled nanosheet devices. Using accurate dissipative DFT-NEGF atomistic-simulation fundamentals and cell layout extrinsics, we demonstrate superior and optimal device characteristics and invertor energy - delays down to sub-30-nm pitches, i.e., a 10 nm scaling boost compared to the nanosheet MOSFET references.

cond-mat.mes-hall↗