arXiv subjects

Jinming Zhuang

Publications and source records attributed to Jinming Zhuang.

12 recordsLinked to original sources

arXiv2026

μ-ORCA: Optimizing Acceleration for Microsecond-Scale Deep Neural Network Inference on ACAP

Heterogeneous reconfigurable platforms with tensor cores, such as AMD ACAP, are increasingly adopted for deep neural network (DNN) inference due to their high throughput and flexibility. However, their suitability for microsecond-scale inference on small problem sizes remains underexplored. In jet-tagging applications in high-energy physics, inefficient on-chip communication and large inter-layer latency prevent existing frameworks from meeting the 1-μs latency budget. Moreover, hardware overheads such as synchronization and VLIW processor prologue are often overlooked, making it infeasible to optimize accelerators correctly. To address these problems, we propose μ-ORCA, a customized heterogeneous accelerator framework for ultra-low-latency model inference. μ-ORCA enables direct inter-layer communication between DNN layers on the AIE array, instead of using shared memory tiles or FPGA fabric. Moreover, a 512-bit/cycle cascade connection is applied instead of a 32-bit/cycle DMA connection. μ-ORCA also provides an overhead-aware performance model that adapts to different NN layer sizes, and conducts design space exploration to optimize end-to-end latency. μ-ORCA supports MLP and DeepSets models with non-MM kernels, including bias, ReLU, and global aggregation on AIE. We evaluate μ-ORCA on the AMD ACAP VEK280 platform. Experimental results show that μ-ORCA achieves average latency reduction of >1.70$\times$ and >1.83$\times$ compared with different state-of-the-art ACAP frameworks, and achieves 0.93 μs latency for a 6-layer real-world DeepSets model, satisfying the latency budget. We open source μ-ORCA at https://github.com/arc-research-lab/u-ORCA.

cs.AR↗

arXiv2026

DORA: Dataflow-Instruction Orchestration Architecture for DNN Acceleration

As deep neural networks develop significantly more diverse and complex, achieving high performance and efficiency on complicated DNN models faces pressing challenges. Modern DNN workloads are increasingly diverse in operation types, tensor shapes, and execution dependencies, making it difficult to sustain high hardware efficiency across models. In addition, a generic accelerator often incurs substantial overhead when executing diverse workloads. To address these problems, we propose DORA, an instruction-based overlay architecture that explicitly describes dataflow via a proposed ISA, enabling fine-grained control of data movement, computation, and synchronization at the layer level. To support flexibility while achieving high performance, DORA adopts a novel on-chip memory management and computation parallelism management mechanism. DORA proposes a compilation framework that can generate instructions for given DNN workloads after a two-stage design space exploration. DORA framework also incorporates a MILP-based and a heuristic-based search engine to generate the schedule solution for different needs and constraints. We prototype DORA on the AMD Versal VCK190 platform, demonstrating its deployability on existing reconfigurable systems. Experimental results show that DORA maintains stable efficiency, with less than 5\% variation on a single vector processor across workloads exhibiting up to 6$\times$ variation in operation counts. Compared to state-of-the-art accelerators, DORA consistently achieves higher performance, delivering up to 5$\times$ throughput improvement. The heuristic-based scheduler further achieves up to 90\% optimality under practical time constraints. DORA is open-sourced at https://github.com/arc-research-lab/DORA.git.

cs.AR↗

arXiv2026

FILCO: Flexible Composing Architecture with Real-Time Reconfigurability for DNN Acceleration

With the development of deep neural network (DNN) enabled applications, achieving high hardware resource efficiency on diverse workloads is non-trivial in heterogeneous computing platforms. Prior works discuss dedicated architectures to achieve maximal resource efficiency. However, a mismatch between hardware and workloads always exists in various diverse workloads. Other works discuss overlay architecture that can dynamically switch dataflow for different workloads. However, these works are still limited by flexibility granularity and induce much resource inefficiency. To solve this problem, we propose a flexible composing architecture, FILCO, that can efficiently match diverse workloads to achieve the optimal storage and computation resource efficiency. FILCO can be reconfigured in real-time and flexibly composed into a unified or multiple independent accelerators. We also propose the FILCO framework, including an analytical model with a two-stage DSE that can achieve the optimal design point. We also evaluate the FILCO framework on the 7nm AMD Versal VCK190 board. Compared with prior works, our design can achieve 1.3x - 5x throughput and hardware efficiency on various diverse workloads.

cs.AR↗

arXiv2026

PHAROS: Pipelined Heterogeneous Accelerators for Real-time Safety-critical Systems With Deadline Compliance

Spatially partitioned heterogeneous accelerators (HAs) are increasingly adopted in embedded systems for their performance and flexibility. Yet most existing HA design frameworks optimize primarily for throughput or quality-of-service (QoS) metrics. They often overlook safety-critical real-time requirements, including hardware support for predictable execution, real-time-aware design space exploration (DSE), and rigorous schedulability analysis. These requirements are essential in safety-critical applications such as smart transportation, where schedulability guarantees directly affect system safety. To address this gap, we present PHAROS, a real-time-centric HA design framework. PHAROS introduces preemption mechanisms and scheduler designs for spatially partitioned HAs under first-in-first-out (FIFO) and earliest-deadline-first (EDF) policies. Leveraging modern real-time theory, we further develop a soft real-time (SRT) schedulability-oriented DSE with objectives and constraints tailored to SRT schedulability. Through comprehensive modeling, analysis, and evaluation across diverse applications, we show that PHAROS's DSE discovers more feasible configurations for a broader range of task sets than throughput-oriented DSE baselines while delivering improved real-time performance. We also provide response-time analyses for the supported scheduling algorithms.

cs.AR↗

arXiv2025

From Loop Nests to Silicon: Mapping AI Workloads onto AMD NPUs with MLIR-AIR

General-purpose compilers abstract away parallelism, locality, and synchronization, limiting their effectiveness on modern spatial architectures. As modern computing architectures increasingly rely on fine-grained control over data movement, execution order, and compute placement for performance, compiler infrastructure must provide explicit mechanisms for orchestrating compute and data to fully exploit such architectures. We introduce MLIR-AIR, a novel, open-source compiler stack built on MLIR that bridges the semantic gap between high-level workloads and fine-grained spatial architectures such as AMD's NPUs. MLIR-AIR defines the AIR dialect, which provides structured representations for asynchronous and hierarchical operations across compute and memory resources. AIR primitives allow the compiler to orchestrate spatial scheduling, distribute computation across hardware regions, and overlap communication with computation without relying on ad hoc runtime coordination or manual scheduling. We demonstrate MLIR-AIR's capabilities through two case studies: matrix multiplication and the multi-head attention block from the LLaMA 2 model. For matrix multiplication, MLIR-AIR achieves up to 78.7% compute efficiency and generates implementations with performance almost identical to state-of-the-art, hand-optimized matrix multiplication written using the lower-level, close-to-metal MLIR-AIE framework. For multi-head attention, we demonstrate that the AIR interface supports fused implementations using approximately 150 lines of code, enabling tractable expression of complex workloads with efficient mapping to spatial hardware. MLIR-AIR transforms high-level structured control flow into spatial programs that efficiently utilize the compute fabric and memory hierarchy of an NPU, leveraging asynchronous execution, tiling, and communication overlap through compiler-managed scheduling.

cs.CL↗

arXiv2025

AGILE: Lightweight and Efficient Asynchronous GPU-SSD Integration

GPUs are critical for compute-intensive applications, yet emerging workloads such as recommender systems, graph analytics, and data analytics often exceed GPU memory capacity. Existing solutions allow GPUs to use CPU DRAM or SSDs as external memory, and the GPU-centric approach enables GPU threads to directly issue NVMe requests, further avoiding CPU intervention. However, current GPU-centric approaches adopt synchronous I/O, forcing threads to stall during long communication delays. We propose AGILE, a lightweight asynchronous GPU-centric I/O library that eliminates deadlock risks and integrates a flexible HBM-based software cache. AGILE overlaps computation and I/O, improving performance by up to 1.88$\times$ across workloads with diverse computation-to-communication ratios. Compared to BaM on DLRM, AGILE achieves up to 1.75$\times$ speedup through efficient design and overlapping; on graph applications, AGILE reduces software cache overhead by up to 3.12$\times$ and NVMe I/O overhead by up to 2.85$\times$; AGILE also lowers per-thread register usage by up to 1.32$\times$.

cs.DC↗

arXiv2024

Challenges and Opportunities to Enable Large-Scale Computing via Heterogeneous Chiplets

Fast-evolving artificial intelligence (AI) algorithms such as large language models have been driving the ever-increasing computing demands in today's data centers. Heterogeneous computing with domain-specific architectures (DSAs) brings many opportunities when scaling up and scaling out the computing system. In particular, heterogeneous chiplet architecture is favored to keep scaling up and scaling out the system as well as to reduce the design complexity and the cost stemming from the traditional monolithic chip design. However, how to interconnect computing resources and orchestrate heterogeneous chiplets is the key to success. In this paper, we first discuss the diversity and evolving demands of different AI workloads. We discuss how chiplet brings better cost efficiency and shorter time to market. Then we discuss the challenges in establishing chiplet interface standards, packaging, and security issues. We further discuss the software programming challenges in chiplet systems.

cs.AR↗

arXiv2024

SSR: Spatial Sequential Hybrid Architecture for Latency Throughput Tradeoff in Transformer Acceleration

With the increase in the computation intensity of the chip, the mismatch between computation layer shapes and the available computation resource significantly limits the utilization of the chip. Driven by this observation, prior works discuss spatial accelerators or dataflow architecture to maximize the throughput. However, using spatial accelerators could potentially increase the execution latency. In this work, we first systematically investigate two execution models: (1) sequentially (temporally) launch one monolithic accelerator, and (2) spatially launch multiple accelerators. From the observations, we find that there is a latency throughput tradeoff between these two execution models, and combining these two strategies together can give us a more efficient latency throughput Pareto front. To achieve this, we propose spatial sequential architecture (SSR) and SSR design automation framework to explore both strategies together when deploying deep learning inference. We use the 7nm AMD Versal ACAP VCK190 board to implement SSR accelerators for four end-to-end transformer-based deep learning models. SSR achieves average throughput gains of 2.53x, 35.71x, and 14.20x under different batch sizes compared to the 8nm Nvidia GPU A10G, 16nm AMD FPGAs ZCU102, and U250. The average energy efficiency gains are 8.51x, 6.75x, and 21.22x, respectively. Compared with the sequential-only solution and spatial-only solution on VCK190, our spatial-sequential-hybrid solutions achieve higher throughput under the same latency requirement and lower latency under the same throughput requirement. We also use SSR analytical models to demonstrate how to use SSR to optimize solutions on other computing platforms, e.g., 14nm Intel Stratix 10 NX.

cs.AR↗

arXiv2023

REFRESH FPGAs: Sustainable FPGA Chiplet Architectures

There is a growing call for greater amounts of increasingly agile computational power for edge and cloud infrastructure to serve the computationally complex needs of ubiquitous computing devices. Thus, an important challenge is addressing the holistic environmental impacts of these next-generation computing systems. To accomplish this, a life-cycle view of sustainability for computing advancements is necessary to reduce environmental impacts such as greenhouse warming gas emissions from these computing choices. Unfortunately, decadal efforts to address operational energy efficiency in computing devices have ignored and in some cases exacerbated embodied impacts from manufacturing these edge and cloud systems, particularly their integrated circuits. During this time FPGA architectures have not changed dramatically except to increase in size. Given this context, we propose REFRESH FPGAs to build new FPGA devices and architectures from recently retired FPGA dies using 2.5D integration. To build REFRESH FPGAs requires creative architectures that leverage existing chiplet pins with an inexpensive to-manufacture interposer coupled with creative design automation. In this paper, we discuss how REFRESH FPGAs can leverage industry trends for renewable energy integration into data centers while providing an overall improvement for sustainability and amortizing their significant embodied cost investment over a much longer ``first'' lifetime.

cs.AR↗

arXiv2023

AIM: Accelerating Arbitrary-precision Integer Multiplication on Heterogeneous Reconfigurable Computing Platform Versal ACAP

Arbitrary-precision integer multiplication is the core kernel of many applications in simulation, cryptography, etc. Existing acceleration of arbitrary-precision integer multiplication includes CPUs, GPUs, FPGAs, and ASICs. Among these accelerators, FPGAs are promised to provide both good energy efficiency and flexibility. Surprisingly, in our implementations, FPGA has the lowest energy efficiency, i.e., 0.29x of the CPU and 0.17x of the GPU with the same generation fabrication. Therefore, key questions arise: Where do the energy efficiency gains of CPUs and GPUs come from? Can reconfigurable computing do better? If can, how to achieve that? We identify that the biggest energy efficiency gains of the CPUs and GPUs come from the dedicated vector units. FPGA uses DSPs and lookup tables to compose the needed computation, which incurs overhead when compared to using vector units directly. New reconfigurable computing, e.g., 'FPGA+vector units' is a novel and feasible solution to improve energy efficiency. In this paper, we propose to map arbitrary-precision integer multiplication onto such a heterogeneous platform, i.e., AMD/Xilinx Versal ACAP architecture. Designing on Versal ACAP incurs several challenges and we propose AIM: Arbitrary-precision Integer Multiplication on Versal ACAP to automate and optimize the design. AIM framework includes design space exploration and AIM automatic code generation to facilitate the system design and verification. We deploy the AIM framework on three different applications, including large integer multiplication (LIM), RSA, and Mandelbrot, on the AMD/Xilinx Versal ACAP VCK190 evaluation board. Our experimental results show that AIM achieves up to 12.6x, and 2.1x energy efficiency gains over the Intel Xeon Ice Lake 6346 CPU, and NVidia A5000 GPU respectively, which brings reconfigurable computing the most energy-efficient platform among CPUs and GPUs.

cs.AR↗

arXiv2023

AutoMM: Energy-Efficient Multi-Data-Type Matrix Multiply Design on Heterogeneous Programmable System-on-Chip

As the increasing complexity of Neural Network(NN) models leads to high demands for computation, AMD introduces a heterogeneous programmable system-on-chip (SoC), i.e., Versal ACAP architectures featured with programmable logic (PL), CPUs, and dedicated AI engines (AIE) ASICs which has a theoretical throughput up to 6.4 TFLOPs for FP32, 25.6 TOPs for INT16 and 102.4 TOPs for INT8. However, the higher level of complexity makes it non-trivial to achieve the theoretical performance even for well-studied applications like matrix-matrix multiply. In this paper, we provide AutoMM, an automatic white-box framework that can systematically generate the design for MM accelerators on Versal which achieves 3.7 TFLOPs, 7.5 TOPs, and 28.2 TOPs for FP32, INT16, and INT8 data type respectively. Our designs are tested on board and achieve gains of 7.20x (FP32), 3.26x (INT16), 6.23x (INT8) energy efficiency than AMD U250 FPGA, 2.32x (FP32) than Nvidia Jetson TX2 GPU, 1.06x (FP32), 1.70x (INT8) than Nvidia A100 GPU.

cs.AR↗

arXiv2023

CHARM: Composing Heterogeneous Accelerators for Matrix Multiply on Versal ACAP Architecture

Dense matrix multiply (MM) serves as one of the most heavily used kernels in deep learning applications. To cope with the high computation demands of these applications, heterogeneous architectures featuring both FPGA and dedicated ASIC accelerators have emerged as promising platforms. For example, the AMD/Xilinx Versal ACAP architecture combines general-purpose CPU cores and programmable logic with AI Engine processors optimized for AI/ML. With 400 AIEs, it provides up to 6.4 TFLOPs performance for 32-bit floating-point data. However, machine learning models often contain both large and small MM operations. While large MM operations can be parallelized efficiently across many cores, small MM operations typically cannot. We observe that executing some small MM layers from the BERT natural language processing model on a large, monolithic MM accelerator in Versal ACAP achieved less than 5% of the theoretical peak performance. Therefore, one key question arises: How can we design accelerators to fully use the abundant computation resources under limited communication bandwidth for applications with multiple MM layers of diverse sizes? We identify the biggest system throughput bottleneck resulting from the mismatch of massive computation resources of one monolithic accelerator and the various MM layers of small sizes in the application. To resolve this problem, we propose the CHARM framework to compose multiple diverse MM accelerator architectures working concurrently towards different layers in one application. We deploy the CHARM framework for four different applications, including BERT, ViT, NCF, MLP, on the AMD Versal ACAP VCK190 evaluation board. Our experiments show that we achieve 1.46 TFLOPs, 1.61 TFLOPs, 1.74 TFLOPs, and 2.94 TFLOPs inference throughput for BERT, ViT, NCF and MLP, which obtain 5.40x, 32.51x, 1.00x and 1.00x throughput gains compared to one monolithic accelerator.

cs.AR↗