Cover Image

FlatAttention: Dataflow and Fabric Collectives Co-Optimization for Efficient Multi-Head Attention on Tile-Based Many-PE Accelerators

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Zhang, Chi, Colagrande, Luca, Andri, Renzo, Benz, Thomas, Islamoglu, Gamze, Nadalini, Alessandro, Conti, Francesco, Li, Yawei, Benini, Luca
Format:	Preprint
Published:	2025
Subjects:	Hardware Architecture
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

FlatAttention: Dataflow and Fabric Collectives Co-Optimization for Large Attention-Based Model Inference on Tile-Based Accelerators
by: Zhang, Chi, et al.
Published: (2026)

A Multicast-Capable AXI Crossbar for Many-core Machine Learning Accelerators
by: Colagrande, Luca, et al.
Published: (2025)

Design in Tiles: Automating GEMM Deployment on Tile-Based Many-PE Accelerators
by: Shen, Aofeng, et al.
Published: (2025)

Toward Attention-based TinyML: A Heterogeneous Accelerated Architecture and Automated Deployment Flow
by: Wiese, Philip, et al.
Published: (2024)

Late Breaking Results: Boosting Efficient Dual-Issue Execution on Lightweight RISC-V Cores
by: Colagrande, Luca, et al.
Published: (2026)

Dual-Issue Execution of Mixed Integer and Floating-Point Workloads on Energy-Efficient In-Order RISC-V Cores
by: Colagrande, Luca, et al.
Published: (2025)

ITA: An Energy-Efficient Attention and Softmax Accelerator for Quantized Transformers
by: İslamoğlu, Gamze, et al.
Published: (2023)

MXDOTP: A RISC-V ISA Extension for Enabling Microscaling (MX) Floating-Point Dot Products
by: İslamoğlu, Gamze, et al.
Published: (2025)

CHIMERA: A Flexible and Scalable 3.1 TOPS/W AI-MCU with Transformer Accelerator and 563 Gb/s Shared-L2 Memory Subsystem with QoS Guarantees
by: Leone, Lorenzo, et al.
Published: (2026)

VMXDOTP: A RISC-V Vector ISA Extension for Efficient Microscaling (MX) Format Acceleration
by: Wipfli, Max, et al.
Published: (2026)

SARIS: Accelerating Stencil Computations on Energy-Efficient RISC-V Compute Clusters with Indirect Stream Registers
by: Scheffler, Paul, et al.
Published: (2024)

VEXP: A Low-Cost RISC-V ISA Extension for Accelerated Softmax Computation in Transformers
by: Wang, Run, et al.
Published: (2025)

Optimizing Offload Performance in Heterogeneous MPSoCs
by: Colagrande, Luca, et al.
Published: (2024)

Taming Offload Overheads in a Massively Parallel Open-Source RISC-V MPSoC: Analysis and Optimization
by: Colagrande, Luca, et al.
Published: (2025)

Late Breaking Results: A RISC-V ISA Extension for Chaining in Scalar Processors
by: Colagrande, Luca, et al.
Published: (2025)

Towards Zero-Stall Matrix Multiplication on Energy-Efficient RISC-V Clusters for Machine Learning Acceleration
by: Colagrande, Luca, et al.
Published: (2025)

A Lightweight High-Throughput Collective-Capable NoC for Large-Scale ML Accelerators
by: Colagrande, Luca, et al.
Published: (2026)

Ara2: Exploring Single- and Multi-Core Vector Processing with an Efficient RVV 1.0 Compliant Open-Source Processor
by: Perotti, Matteo, et al.
Published: (2023)

Stella Nera: A Differentiable Maddness-Based Hardware Accelerator for Efficient Approximate Matrix Multiplication
by: Schönleber, Jannis, et al.
Published: (2023)

Fused-Tiled Layers: Minimizing Data Movement on RISC-V SoCs with Software-Managed Caches
by: Jung, Victor J. B., et al.
Published: (2025)

HyperCroc: End-to-End Open-Source RISC-V MCU with a Plug-In Interface for Domain-Specific Accelerators
by: Sauter, Philippe, et al.
Published: (2026)

A "New Ara" for Vector Computing: An Open Source Highly Efficient RISC-V V 1.0 Vector Processor Design
by: Perotti, Matteo, et al.
Published: (2022)

Multilayer Dataflow: Orchestrate Butterfly Sparsity to Accelerate Attention Computation
by: Wu, Haibin, et al.
Published: (2024)

FractalSync: Lightweight Scalable Global Synchronization of Massive Bulk Synchronous Parallel AI Accelerators
by: Isachi, Victor, et al.
Published: (2025)

A Direct Memory Access Controller (DMAC) for Irregular Data Transfers on RISC-V Linux Systems
by: Benz, Thomas, et al.
Published: (2025)

Occamy: A 432-Core 28.1 DP-GFLOP/s/W 83% FPU Utilization Dual-Chiplet, Dual-HBM2E RISC-V-based Accelerator for Stencil and Sparse Linear Algebra Computations with 8-to-64-bit Floating-Point Support in 12nm FinFET
by: Paulin, Gianna, et al.
Published: (2024)

Towards Reliable Systems: A Scalable Approach to AXI4 Transaction Monitoring
by: Liang, Chaoqun, et al.
Published: (2025)

ControlPULPlet: A Flexible Real-time Multi-core RISC-V Controller for 2.5D Systems-in-package
by: Ottaviano, Alessandro, et al.
Published: (2024)

A Flexible Template for Edge Generative AI with High-Accuracy Accelerated Softmax & GELU
by: Belano, Andrea, et al.
Published: (2024)

Fast Cross-Operator Optimization of Attention Dataflow
by: Chang, Haodong, et al.
Published: (2026)

Optimizing Foundation Model Inference on a Many-tiny-core Open-source RISC-V Platform
by: Potocnik, Viviane, et al.
Published: (2024)

SentryCore: A RISC-V Co-Processor System for Safe, Real-Time Control Applications
by: Rogenmoser, Michael, et al.
Published: (2024)

A Gigabit, DMA-enhanced Open-Source Ethernet Controller for Mixed-Criticality Systems
by: Liang, Chaoqun, et al.
Published: (2024)

FlooNoC: A 645 Gbps/link 0.15 pJ/B/hop Open-Source NoC with Wide Physical Links and End-to-End AXI4 Parallel Multi-Stream Support
by: Fischer, Tim, et al.
Published: (2024)

Insights from Basilisk: Are Open-Source EDA Tools Ready for a Multi-Million-Gate, Linux-Booting RV64 SoC Design?
by: Sauter, Philippe, et al.
Published: (2024)

Basilisk: An End-to-End Open-Source Linux-Capable RISC-V SoC in 130nm CMOS
by: Scheffler, Paul, et al.
Published: (2024)

Occamy: A 432-Core Dual-Chiplet Dual-HBM2E 768-DP-GFLOP/s RISC-V System for 8-to-64-bit Dense and Sparse Computing in 12nm FinFET
by: Scheffler, Paul, et al.
Published: (2025)

Implementing and Optimizing the Scaled Dot-Product Attention on Streaming Dataflow
by: Sohn, Gina, et al.
Published: (2024)

A Persistent-State Dataflow Accelerator for Memory-Bound Linear Attention Decode on FPGA
by: Gupta, Neelesh, et al.
Published: (2026)

Toward Open-Source Chiplets for HPC and AI: Occamy and Beyond
by: Scheffler, Paul, et al.
Published: (2025)