Cover Image

Stratum: System-Hardware Co-Design with Tiered Monolithic 3D-Stackable DRAM for Efficient MoE Serving

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Pan, Yue, Xia, Zihan, Hsu, Po-Kai, Hu, Lanxiang, Kim, Hyungyo, Sharda, Janak, Zhou, Minxuan, Kim, Nam Sung, Yu, Shimeng, Rosing, Tajana, Kang, Mingu
Format:	Preprint
Published:	2025
Subjects:	Hardware Architecture Emerging Technologies Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

PIM-FW: Hardware-Software Co-Design of All-pairs Shortest Paths in DRAM
by: Lu, Tsung-Han, et al.
Published: (2025)

GenDRAM:Hardware-Software Co-Design of General Platform in DRAM
by: Lu, Tsung-Han, et al.
Published: (2026)

Hybrid SLC-MLC RRAM Mixed-Signal Processing-in-Memory Architecture for Transformer Acceleration via Gradient Redistribution
by: Song, Chang Eun, et al.
Published: (2025)

Row Hammer Effect and Floating Body Effect of Monolithic 3D Stackable 1T1C DRAM
by: Cho, Sungwon, et al.
Published: (2025)

Proxima: Near-storage Acceleration for Graph-based Approximate Nearest Neighbor Search in 3D NAND
by: Xu, Weihong, et al.
Published: (2023)

Fast-OverlaPIM: A Fast Overlap-driven Mapping Framework for Processing In-Memory Neural Network Acceleration
by: Wang, Xuan, et al.
Published: (2024)

HAVEN: High-Bandwidth Flash Augmented Vector Engine for Large-Scale Approximate Nearest-Neighbor Search Acceleration
by: Hsu, Po-Kai, et al.
Published: (2026)

SLIM: A Heterogeneous Accelerator for Edge Inference of Sparse Large Language Model via Adaptive Thresholding
by: Xu, Weihong, et al.
Published: (2025)

A3D-MoE: Acceleration of Large Language Models with Mixture of Experts via 3D Heterogeneous Integration
by: Huang, Wei-Hsing, et al.
Published: (2025)

FaTRQ: Tiered Residual Quantization for LLM Vector Search in Far-Memory-Aware ANNS Systems
by: Zhang, Tianqi, et al.
Published: (2026)

System-Technology Co-Optimization of Bitline Routing and Bonding Pathways in Monolithic 3D DRAM Architectures
by: Lee, Kiseok, et al.
Published: (2026)

Hardware-Software Co-design for 3D-DRAM-based LLM Serving Accelerator
by: Li, Cong, et al.
Published: (2026)

DRAMScope: Uncovering DRAM Microarchitecture and Characteristics by Issuing Memory Commands
by: Nam, Hwayong, et al.
Published: (2024)

SpANNS: Optimizing Approximate Nearest Neighbor Search for Sparse Vectors Using Near Memory Processing
by: Zhang, Tianqi, et al.
Published: (2026)

RAPID-Graph: Recursive All-Pairs Shortest Paths Using Processing-in-Memory for Dynamic Programming on Graphs
by: Chen, Yanru, et al.
Published: (2025)

Low-Power Encoding for PAM-3 DRAM Bus
by: Nam, Jonghyeon, et al.
Published: (2024)

Sangam: Chiplet-Based DRAM-PIM Accelerator with CXL Integration for LLM Inferencing
by: Kiyawat, Khyati, et al.
Published: (2025)

Rethinking LLM Inference Bottlenecks: Insights from Latent Attention and Mixture-of-Experts
by: Yun, Sungmin, et al.
Published: (2025)

SliceMoE: Bit-Sliced Expert Caching under Miss-Rate Constraints for Efficient MoE Inference
by: Choi, Yuseon, et al.
Published: (2025)

FSL-HDnn: A 5.7 TOPS/W End-to-end Few-shot Learning Classifier Accelerator with Feature Extraction and Hyperdimensional Computing
by: Yang, Haichao, et al.
Published: (2024)

ATiM: Autotuning Tensor Programs for Processing-in-DRAM
by: Shin, Yongwon, et al.
Published: (2024)

FeNOMS: Enhancing Open Modification Spectral Library Search with In-Storage Processing on Ferroelectric NAND (FeNAND) Flash
by: Pinge, Sumukh, et al.
Published: (2025)

Efficient Open Modification Spectral Library Searching in High-Dimensional Space with Multi-Level-Cell Memory
by: Fan, Keming, et al.
Published: (2024)

FSL-HDnn: A 40 nm Few-shot On-Device Learning Accelerator with Integrated Feature Extraction and Hyperdimensional Computing
by: Xu, Weihong, et al.
Published: (2025)

LOCALUT: Harnessing Capacity-Computation Tradeoffs for LUT-Based Inference in DRAM-PIM
by: Hong, Junguk, et al.
Published: (2026)

SpecPCM: A Low-power PCM-based In-Memory Computing Accelerator for Full-stack Mass Spectrometry Analysis
by: Fan, Keming, et al.
Published: (2024)

MoE-Hub: Taming Software Complexity for Seamless MoE Overlap with Hardware-Accelerated Communication on Multi-GPU Systems
by: Zhou, Zhuoshan, et al.
Published: (2026)

RED: Energy Optimization Framework for eDRAM-based PIM with Reconfigurable Voltage Swing and Retention-aware Scheduling
by: Kim, Jae-Young, et al.
Published: (2025)

Shifting in-DRAM
by: Tegge, William C., et al.
Published: (2026)

Bandwidth-Effective DRAM Cache for GPUs with Storage-Class Memory
by: Hong, Jeongmin, et al.
Published: (2024)

Efficient MoE Serving in the Memory-Bound Regime: Balance Activated Experts, Not Tokens
by: Yu, Yanpeng, et al.
Published: (2025)

CHIME: Chiplet-based Heterogeneous Near-Memory Acceleration for Edge Multimodal LLM Inference
by: Chen, Yanru, et al.
Published: (2025)

NeoMem: Hardware/Software Co-Design for CXL-Native Memory Tiering
by: Zhou, Zhe, et al.
Published: (2024)

ELMoE-3D: Leveraging Intrinsic Elasticity of MoE for Hybrid-Bonding-Enabled Self-Speculative Decoding in On-Premises Serving
by: Choi, Yuseon, et al.
Published: (2026)

SPADE: Sparse Pillar-based 3D Object Detection Accelerator for Autonomous Driving
by: Lee, Minjae, et al.
Published: (2023)

Kelle: Co-design KV Caching and eDRAM for Efficient LLM Serving in Edge Computing
by: Xia, Tianhua, et al.
Published: (2025)

MoE-GPS: Guidlines for Prediction Strategy for Dynamic Expert Duplication in MoE Load Balancing
by: Ma, Haiyue, et al.
Published: (2025)

GEN-Graph: Heterogeneous PIM Accelerator for General Computational Patterns in Graph-based Dynamic Programming
by: Chen, Yanru, et al.
Published: (2026)

Sectored DRAM: A Practical Energy-Efficient and High-Performance Fine-Grained DRAM Architecture
by: Olgun, Ataberk, et al.
Published: (2022)

EasyDRAM: An FPGA-based Infrastructure for Fast and Accurate End-to-End Evaluation of Emerging DRAM Techniques
by: Canpolat, Oğuzhan, et al.
Published: (2025)