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
This content will become publicly available on February 12, 2024
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 (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
- NSF-PAR ID:
- 10432083
- Date Published:
- Journal Name:
- Proceedings of the 2023 ACM/SIGDA International Symposium on Field Programmable Gate Arrays
- Page Range / eLocation ID:
- 153 to 164
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Graph Neural Networks (GNNs) have drawn tremendous attention due to their unique capability to extend Machine Learning (ML) approaches to applications broadly-defined as having unstructured data, especially graphs. Compared with other Machine Learning (ML) modalities, the acceleration of Graph Neural Networks (GNNs) is more challenging due to the irregularity and heterogeneity derived from graph typologies. Existing efforts, however, have focused mainly on handling graphs’ irregularity and have not studied their heterogeneity. To this end we propose H-GCN, a PL (Programmable Logic) and AIE (AI Engine) based hybrid accelerator that leverages the emerging heterogeneity of Xilinx Versal Adaptive Compute Acceleration Platforms (ACAPs) to achieve high-performance GNN inference. In particular, H-GCN partitions each graph into three subgraphs based on its inherent heterogeneity, and processes them using PL and AIE, respectively. To further improve performance, we explore the sparsity support of AIE and develop an efficient density-aware method to automatically map tiles of sparse matrix-matrix multiplication (SpMM) onto the systolic tensor array. Compared with state-of-the-art GCN accelerators, H-GCN achieves, on average, speedups of 1.1∼2.3×.more » « less
-
Graph processing recently received intensive interests in light of a wide range of needs to understand relationships. It is well-known for the poor locality and high memory bandwidth requirement. In conventional architectures, they incur a significant amount of data movements and energy consumption which motivates several hardware graph processing accelerators. The current graph processing accelerators rely on memory access optimizations or placing computation logics close to memory. Distinct from all existing approaches, we leverage an emerging memory technology to accelerate graph processing with analog computation. This paper presents GRAPHR, the first ReRAM-based graph processing accelerator. GRAPHR follows the principle of near-data processing and explores the opportunity of performing massive parallel analog operations with low hardware and energy cost. The analog computation is suitable for graph processing because: 1) The algorithms are iterative and could inherently tolerate the imprecision; 2) Both probability calculation (e.g., PageRank and Collaborative Filtering) and typical graph algorithms involving integers (e.g., BFS/SSSP) are resilient to errors. The key insight of GRAPHR is that if a vertex program of a graph algorithm can be expressed in sparse matrix vector multiplication (SpMV), it can be efficiently performed by ReRAM crossbar. We show that this assumption is generally true for a large set of graph algorithms. GRAPHR is a novel accelerator architecture consisting of two components: memory ReRAM and graph engine (GE). The core graph computations are performed in sparse matrix format in GEs (ReRAM crossbars). The vector/matrix-based graph computation is not new, but ReRAM offers the unique opportunity to realize the massive parallelism with unprecedented energy efficiency and low hardware cost. With small subgraphs processed by GEs, the gain of performing parallel operations overshadows the wastes due to sparsity. The experiment results show that GRAPHR achieves a 16.01X (up to 132.67X) speedup and a 33.82X energy saving on geometric mean compared to a CPU baseline system. Compared to GPU, GRAPHR achieves 1.69X to 2.19X speedup and consumes 4.77X to 8.91X less energy. GRAPHR gains a speedup of 1.16X to 4.12X, and is 3.67X to 10.96X more energy efficiency compared to PIM-based architecture.more » « less
-
Spiking neural networks (SNNs) offer a promising biologically-plausible computing model and lend themselves to ultra-low-power event-driven processing on neuromorphic processors. Compared with the conventional artificial neural networks, SNNs are well-suited for processing complex spatiotemporal data. Despite its significance, dataflow optimization of spiking neural accelerator architectures has not been extensively studied. Recognizing the need for efficient processing of complex spatiotemporal data while considering the all-or-none nature of spiking activities, we propose holistic reconfigurable dataflow optimization for systolic array acceleration of spiking convolutional networks (S-CNNs). A novel scheme is introduced for parallel acceleration of computation across multiple time points, which further allows for systemic optimization of variable tiling for a large performance and efficiency gains. We show how variable tiling, in particular, the positioning of the temporal dimension, can be targeted to optimize data movement, throughput, and energy efficiency. Furthermore, we explore joint layer-dependent dataflow and accelerator hardware optimization to further boost performance and energy efficiency. To support systemic design space exploration, we develop an SNN dataflow simulator capable of analyzing the throughput and energy dissipation of systolic array accelerators for any targeted S-CNN while considering the inherent spatiotemporal characteristics of spiking neural computation. The proposed techniques deliver orders of magnitude of improvements on throughput, energy efficiency, and delay-energy product for accelerating deep Alexnet and VGG-16 SNNs.more » « less
-
Matrix multiplication is one of the bottleneck computations for training the weights within deep neural networks. To speed up the training phase, we propose to use faster algorithms for matrix multiplication known as Arbitrary Precision Approximating (APA) algorithms. APA algorithms perform asymptotically fewer arithmetic operations than the classical algorithm, but they compute an approximate result with an error that can be made arbitrarily small in exact arithmetic. Practical APA algorithms provide significant reduction in computation time and still provide enough accuracy for many applications like neural network training. We demonstrate that APA algorithms can be efficiently implemented and parallelized for multicore CPUs to obtain up to 28% and 21% speedups over the fastest implementation of the classical algorithm using one core and 12 cores, respectively. Furthermore, using these algorithms to train a Multi-Layer Perceptron (MLP) network yields no significant reduction in the training or testing error. Our performance results on a large MLP network show overall sequential and multithreaded performance improvements of up to 25% and 13%, respectively. We also demonstrate up to 15% improvement when training the fully connected layers of the VGG-19 image classification network.more » « less
