arXiv subjects

Fanchen Kong

Publications and source records attributed to Fanchen Kong.

5 recordsLinked to original sources

arXiv2026

A 16 nm 1.60TOPS/W High Utilization DNN Accelerator with 3D Spatial Data Reuse and Efficient Shared Memory Access

Achieving high compute utilization across a wide range of AI workloads is crucial for the efficiency of versatile DNN accelerators. This paper presents the Voltra chip and its utilization-optimised DNN accelerator architecture, which leverages 3-Dimensional (3D) spatial data reuse along with efficient and flexible shared memory access. The 3D spatial dataflow enables balanced spatial data reuse across three dimensions, improving spatial utilization by up to 2.0x compared to a conventional 2D design. Inside the shared memory access architecture, Voltra incorporates flexible data streamers that enable mixed-grained hardware data pre-fetching and dynamic memory allocation, further improving the temporal utilization by 2.12-2.94x and achieving 1.15-2.36x total latency speedup compared with the non-prefetching and separated memory architecture, respectively. Fabricated in 16nm technology, our chip achieves 1.60 TOPS/W peak system energy efficiency and 1.25 TOPS/mm2 system area efficiency, which is competitive with state-of-the-art solutions while achieving high utilization across diverse workloads.

cs.AR↗

arXiv2025

Torrent: A Distributed DMA for Efficient and Flexible Point-to-Multipoint Data Movement

The growing disparity between computational power and on-chip communication bandwidth is a critical bottleneck in modern Systems-on-Chip (SoCs), especially for data-parallel workloads like AI. Efficient point-to-multipoint (P2MP) data movement, such as multicast, is essential for high performance. However, native multicast support is lacking in standard interconnect protocols. Existing P2MP solutions, such as multicast-capable Network-on-Chip (NoC), impose additional overhead to the network hardware and require modifications to the interconnect protocol, compromising scalability and compatibility. This paper introduces Torrent, a novel distributed DMA architecture that enables efficient P2MP data transfers without modifying NoC hardware and interconnect protocol. Torrent conducts P2MP data transfers by forming logical chains over the NoC, where the data traverses through targeted destinations resembling a linked list. This Chainwrite mechanism preserves the P2P nature of every data transfer while enabling flexible data transfers to an unlimited number of destinations. To optimize the performance and energy consumption of Chainwrite, two scheduling algorithms are developed to determine the optimal chain order based on NoC topology. Our RTL and FPGA prototype evaluations using both synthetic and real workloads demonstrate significant advantages in performance, flexibility, and scalability over network-layer multicast. Compared to the unicast baseline, Torrent achieves up to a 7.88x speedup. ASIC synthesis on 16nm technology confirms the architecture's minimal footprint in area (1.2%) and power (2.3%). Thanks to the Chainwrite, Torrent delivers scalable P2MP data transfers with a small cycle overhead of 82CC and area overhead of 207um2 per destination.

cs.AR↗

arXiv2025

XDMA: A Distributed, Extensible DMA Architecture for Layout-Flexible Data Movements in Heterogeneous Multi-Accelerator SoCs

As modern AI workloads increasingly rely on heterogeneous accelerators, ensuring high-bandwidth and layout-flexible data movements between accelerator memories has become a pressing challenge. Direct Memory Access (DMA) engines promise high bandwidth utilization for data movements but are typically optimal only for contiguous memory access, thus requiring additional software loops for data layout transformations. This, in turn, leads to excessive control overhead and underutilized on-chip interconnects. To overcome this inefficiency, we present XDMA, a distributed and extensible DMA architecture that enables layout-flexible data movements with high link utilization. We introduce three key innovations: (1) a data streaming engine as XDMA Frontend, replacing software address generators with hardware ones; (2) a distributed DMA architecture that maximizes link utilization and separates configuration from data transfer; (3) flexible plugins for XDMA enabling on-the-fly data manipulation during data transfers. XDMA demonstrates up to 151.2x/8.2x higher link utilization than software-based implementations in synthetic workloads and achieves 2.3x average speedup over accelerators with SoTA DMA in real-world applications. Our design incurs <2% area overhead over SoTA DMA solutions while consuming 17% of system power. XDMA proves that co-optimizing memory access, layout transformation, and interconnect protocols is key to unlocking heterogeneous multi-accelerator SoC performance.

cs.AR↗

arXiv2025

OptiCorNet: Optimizing Sequence-Based Context Correlation for Visual Place Recognition

Visual Place Recognition (VPR) in dynamic and perceptually aliased environments remains a fundamental challenge for long-term localization. Existing deep learning-based solutions predominantly focus on single-frame embeddings, neglecting the temporal coherence present in image sequences. This paper presents OptiCorNet, a novel sequence modeling framework that unifies spatial feature extraction and temporal differencing into a differentiable, end-to-end trainable module. Central to our approach is a lightweight 1D convolutional encoder combined with a learnable differential temporal operator, termed Differentiable Sequence Delta (DSD), which jointly captures short-term spatial context and long-range temporal transitions. The DSD module models directional differences across sequences via a fixed-weight differencing kernel, followed by an LSTM-based refinement and optional residual projection, yielding compact, discriminative descriptors robust to viewpoint and appearance shifts. To further enhance inter-class separability, we incorporate a quadruplet loss that optimizes both positive alignment and multi-negative divergence within each batch. Unlike prior VPR methods that treat temporal aggregation as post-processing, OptiCorNet learns sequence-level embeddings directly, enabling more effective end-to-end place recognition. Comprehensive evaluations on multiple public benchmarks demonstrate that our approach outperforms state-of-the-art baselines under challenging seasonal and viewpoint variations.

cs.CV↗

arXiv2025

DataMaestro: A Versatile and Efficient Data Streaming Engine Bringing Decoupled Memory Access To Dataflow Accelerators

Deep Neural Networks (DNNs) have achieved remarkable success across various intelligent tasks but encounter performance and energy challenges in inference execution due to data movement bottlenecks. We introduce DataMaestro, a versatile and efficient data streaming unit that brings the decoupled access/execute architecture to DNN dataflow accelerators to address this issue. DataMaestro supports flexible and programmable access patterns to accommodate diverse workload types and dataflows, incorporates fine-grained prefetch and addressing mode switching to mitigate bank conflicts, and enables customizable on-the-fly data manipulation to reduce memory footprints and access counts. We integrate five DataMaestros with a Tensor Core-like GeMM accelerator and a Quantization accelerator into a RISC-V host system for evaluation. The FPGA prototype and VLSI synthesis results demonstrate that DataMaestro helps the GeMM core achieve nearly 100% utilization, which is 1.05-21.39x better than state-of-the-art solutions, while minimizing area and energy consumption to merely 6.43% and 15.06% of the total system.

cs.AR↗