An official website of the United States government
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. An array of 400 AI Engine processors executing at 1 GHz can provide up to 6.4 TFLOPS performance for 32-bit floating-point (FP32) 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 end-to-end applications with multiple MM layers of diverse sizes? We identify the biggest system throughput bottleneck resulting from the mismatch between 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 composemultiple diverse MM accelerator architecturesworking concurrently on different layers within one application. CHARM includes analytical models which guide design space exploration to determine accelerator partitions and layer scheduling. To facilitate system designs, CHARM automatically generates code, enabling thorough onboard design verification. We deploy the CHARM framework on four different deep learning applications in FP32, INT16, and INT8 data types, including BERT, ViT, NCF, and MLP, on the AMD/Xilinx 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 in FP32 data type, respectively, which obtain 5.29\(\times\), 32.51\(\times\), 1.00\(\times\), and 1.00\(\times\)throughput gains compared to one monolithic accelerator. CHARM achieves the maximum throughput of 1.91 TOPS, 1.18 TOPS, 4.06 TOPS, and 5.81 TOPS in the INT16 data type for the four applications. The maximum throughput achieved by CHARM in the INT8 data type is 3.65 TOPS, 1.28 TOPS, 10.19 TOPS, and 21.58 TOPS, respectively. We have open-sourced our tools, including detailed step-by-step guides to reproduce all the results presented in this paper and to enable other users to learn and leverage CHARM framework and tools in their end-to-end systems:https://github.com/arc-research-lab/CHARM.
more »
« less
High Performance, Low Power Matrix Multiply Design on ACAP: from Architecture, Design Challenges and DSE Perspectives
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, 2.32x (FP32) than Nvidia Jetson TX2, 1.06x (FP32), 1.70x (INT8) than Nvidia A100.
more »
« less
- PAR ID:
- 10464676
- Date Published:
- Journal Name:
- 2023 60th ACM/IEEE Design Automation Conference (DAC)
- Page Range / eLocation ID:
- 1 to 6
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
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 (PL) with AI Engine processors (AIE) optimized for AI/ML. An array of 400 AI Engine processors executing at 1 GHz can theoretically provide up to 6.4 TFLOPs performance for 32-bit floating-point (fp32) 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. In our investigation, 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 end-to-end 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 within one application. CHARM includes analytical models which guide design space exploration to determine accelerator partitions and layer scheduling. To facilitate the system designs, CHARM automatically generates code, enabling thorough onboard design verification. We deploy the CHARM framework for four different deep learning applications, including BERT, ViT, NCF, MLP, on the AMD/Xilinx 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, MLP, respectively, which obtain 5.40x, 32.51x, 1.00x and 1.00x throughput gains compared to one monolithic accelerator.more » « less
-
As AI continues to grow, modern applications are becoming more data- and compute-intensive, driving the development of specialized AI chips to meet these demands. One example is AMD's AI Engine (AIE), a dedicated hardware system that includes a 2D array of high-frequency very-long instruction words (VLIW) vector processors to provide high computational throughput and reconfigurability. However, AIE's specialized architecture presents tremendous challenges in programming and compiler optimization. Existing AIE programming frameworks lack a clean abstraction to represent multi-level parallelism in AIE; programmers have to figure out the parallelism within a kernel, manually do the partition, and assign sub-tasks to different AIE cores to exploit parallelism. These significantly lower the programming productivity. Furthermore, some AIE architectures include FPGAs to provide extra flexibility, but there is no unified intermediate representation (IR) that captures these architectural differences. As a result, existing compilers can only optimize the AIE portions of the code, overlooking potential FPGA bottlenecks and leading to suboptimal performance. To address these limitations, we introduce ARIES, an agile multi-level intermediate representation (MLIR) based compilation flow for reconfigurable devices with AIEs. ARIES introduces a novel programming model that allows users to map kernels to separate AIE cores, exploiting task- and tile-level parallelism without restructuring code. It also includes a declarative scheduling interface to explore instruction-level parallelism within each core. At the IR level, we propose a unified MLIR-based representation for AIE architectures, both with or without FPGA, facilitating holistic optimization and better portability across AIE device families. For the General Matrix Multiply (GEMM) benchmark, ARIES achieves 4.92 TFLOPS, 15.86 TOPS, and 45.94 TOPS throughput under FP32, INT16, and, INT8 data types on Versal VCK190 respectively. Compared with the state-of-the-art (SOTA) work CHARM for AIE, ARIES improves the throughput by 1.17x, 1.59x, and 1.47x correspondingly. For ResNet residual layer, ARIES achieves up to 22.58x speedup compared with optimized SOTA work Riallto on Ryzen-AI NPU. ARIES is open-sourced on GitHub: https://github.com/arc-research-lab/Aries.more » « less
-
While Vision Transformers (ViTs) have shown consistent progress in computer vision, deploying them for real-time decision-making scenarios (< 1 ms) is challenging. Current computing platforms like CPUs, GPUs, or FPGA-based solutions struggle to meet this deterministic low-latency real-time requirement, even with quantized ViT models. Some approaches use pruning or sparsity to reduce model size and latency, but this often results in accuracy loss. To address the aforementioned constraints, in this work, we propose EQ-ViT, an end-to-end acceleration framework with novel algorithm and architecture co-design features to enable real-time ViT acceleration on AMD Versal Adaptive Compute Acceleration Platform (ACAP). The contributions are four-fold. First, we perform in-depth kernel- level performance profiling & analysis and explain the bottlenecks for existing acceleration solutions on GPU, FPGA, and ACAP. Second, on the hardware level, we introduce a new spatial and heterogeneous accelerator architecture, EQ-ViT architec- ture. This architecture leverages the heterogeneous features of ACAP, where both FPGA and artificial intelligence engines (AIEs) coexist on the same system-on-chip (SoC). Third, On the algorithm level, we create a comprehensive quantization-aware training strategy, EQ-ViT algorithm. This strategy concurrently quantizes both weights and activations into 8-bit integers, aiming to improve accuracy rather than compromise it during quanti- zation. Notably, the method also quantizes nonlinear functions for efficient hardware implementation. Fourth, we design EQ- ViT automation framework to implement the EQ-ViT architec- ture for four different ViT applications on the AMD Versal ACAP VCK190 board, achieving accuracy improvement with 2.4%, and average speedups of 315.0x, 3.39x, 3.38x, 14.92x, 59.5x, 13.1x over computing solutions of Intel Xeon 8375C vCPU, Nvidia A10G, A100, Jetson AGX Orin GPUs, and AMD ZCU102, U250 FPGAs. The energy efficiency gains are 62.2x, 15.33x, 12.82x, 13.31x, 13.5x, 21.9x.more » « less
-
Due to the recent announcement of the Frontier supercomputer, many scientific application developers are working to make their applications compatible with AMD (CPU-GPU) architectures, which means moving away from the traditional CPU and NVIDIA-GPU systems. Due to the current limitations of profiling tools for AMD GPUs, this shift leaves a void in how to measure application performance on AMD GPUs. In this article, we design an instruction roofline model for AMD GPUs using AMD’s ROCProfiler and a benchmarking tool, BabelStream (the HIP implementation), as a way to measure an application’s performance in instructions and memory transactions on new AMD hardware. Specifically, we create instruction roofline models for a case study scientific application, PIConGPU, an open source particle-in-cell simulations application used for plasma and laser-plasma physics on the NVIDIA V100, AMD Radeon Instinct MI60, and AMD Instinct MI100 GPUs. When looking at the performance of multiple kernels of interest in PIConGPU we find that although the AMD MI100 GPU achieves a similar, or better, execution time compared to the NVIDIA V100 GPU, profiling tool differences make comparing performance of these two architectures hard. When looking at execution time, GIPS, and instruction intensity, the AMD MI60 achieves the worst performance out of the three GPUs used in this work.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1109/DAC56929.2023.10247981
