اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدWoodpecker-DL: Accelerating Deep Neural Networks via Hardware-Aware Multifaceted Optimizations
93
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
Accelerating deep model training and inference is crucial in practice. Existing deep learning frameworks usually concentrate on optimizing training speed and pay fewer attentions to inference-specific optimizations. Actually, model inference differs from training in terms of computation, e.g. parameters are refreshed each gradient update step during training, but kept invariant during inference. These special characteristics of model inference open new opportunities for its optimization. In this paper, we propose a hardware-aware optimization framework, namely Woodpecker-DL (WPK), to accelerate inference by taking advantage of multiple joint optimizations from the perspectives of graph optimization, automated searches, domain-specific language (DSL) compiler techniques and system-level exploration. In WPK, we investigated two new automated search approaches based on genetic algorithm and reinforcement learning, respectively, to hunt the best operator code configurations targeting specific hardware. A customized DSL compiler is further attached to these search algorithms to generate efficient codes. To create an optimized inference plan, WPK systematically explores high-speed operator implementations from third-party libraries besides our automatically generated codes and singles out the best implementation per operator for use. Extensive experiments demonstrated that on a Tesla P100 GPU, we can achieve the maximum speedup of 5.40 over cuDNN and 1.63 over TVM on individual convolution operators, and run up to 1.18 times faster than TensorRT for end-to-end model inference.
قيم البحث
اقرأ أيضاً
Industrial Internet of Things (IIoT) applications can benefit from leveraging edge computing. For example, applications underpinned by deep neural networks (DNN) models can be sliced and distributed across the IIoT device and the edge of the network
for improving the overall performance of inference and for enhancing privacy of the input data, such as industrial product images. However, low network performance between IIoT devices and the edge is often a bottleneck. In this study, we develop ScissionLite, a holistic framework for accelerating distributed DNN inference using the Transfer Layer (TL). The TL is a traffic-aware layer inserted between the optimal slicing point of a DNN model slice in order to decrease the outbound network traffic without a significant accuracy drop. For the TL, we implement a new lightweight down/upsampling network for performance-limited IIoT devices. In ScissionLite, we develop ScissionTL, the Preprocessor, and the Offloader for end-to-end activities for deploying DNN slices with the TL. They decide the optimal slicing point of the DNN, prepare pre-trained DNN slices including the TL, and execute the DNN slices on an IIoT device and the edge. Employing the TL for the sliced DNN models has a negligible overhead. ScissionLite improves the inference latency by up to 16 and 2.8 times when compared to execution on the local device and an existing state-of-the-art model slicing approach respectively.
Partitioning and distributing deep neural networks (DNNs) across end-devices, edge resources and the cloud has a potential twofold advantage: preserving privacy of the input data, and reducing the ingress bandwidth demand beyond the edge. However, fo
r a given DNN, identifying the optimal partition configuration for distributing the DNN that maximizes performance is a significant challenge. This is because the combination of potential target hardware resources that maximizes performance and the sequence of layers of the DNN that should be distributed across the target resources needs to be determined, while accounting for user-defined objectives/constraints for partitioning. This paper presents Scission, a tool for automated benchmarking of DNNs on a given set of target device, edge and cloud resources for determining optimal partitions that maximize DNN performance. The decision-making approach is context-aware by capitalizing on hardware capabilities of the target resources, their locality, the characteristics of DNN layers, and the network condition. Experimental studies are carried out on 18 DNNs. The decisions made by Scission cannot be manually made by a human given the complexity and the number of dimensions affecting the search space. The benchmarking overheads of Scission allow for responding to operational changes periodically rather than in real-time. Scission is available for public download at https://github.com/qub-blesson/Scission.
Network pruning can reduce the high computation cost of deep neural network (DNN) models. However, to maintain their accuracies, sparse models often carry randomly-distributed weights, leading to irregular computations. Consequently, sparse models ca
nnot achieve meaningful speedup on commodity hardware (e.g., GPU) built for dense matrix computations. As such, prior works usually modify or design completely new sparsity-optimized architectures for exploiting sparsity. We propose an algorithm-software co-designed pruning method that achieves latency speedups on existing dense architectures. Our work builds upon the insight that the matrix multiplication generally breaks the large matrix into multiple smaller tiles for parallel execution. We propose a tiling-friendly tile-wise sparsity pattern, which maintains a regular pattern at the tile level for efficient execution but allows for irregular, arbitrary pruning at the global scale to maintain the high accuracy. We implement and evaluate the sparsity pattern on GPU tensor core, achieving a 1.95x speedup over the dense model.
As neural network model sizes have dramatically increased, so has the interest in various techniques to reduce their parameter counts and accelerate their execution. An active area of research in this field is sparsity - encouraging zero values in pa
rameters that can then be discarded from storage or computations. While most research focuses on high levels of sparsity, there are challenges in universally maintaining model accuracy as well as achieving significant speedups over modern matrix-math hardware. To make sparsity adoption practical, the NVIDIA Ampere GPU architecture introduces sparsity support in its matrix-math units, Tensor Cores. We present the design and behavior of Sparse Tensor Cores, which exploit a 2:4 (50%) sparsity pattern that leads to twice the math throughput of dense matrix units. We also describe a simple workflow for training networks that both satisfy 2:4 sparsity pattern requirements and maintain accuracy, verifying it on a wide range of common tasks and model architectures. This workflow makes it easy to prepare accurate models for efficient deployment on Sparse Tensor Cores.
Designing accurate and efficient convolutional neural architectures for vast amount of hardware is challenging because hardware designs are complex and diverse. This paper addresses the hardware diversity challenge in Neural Architecture Search (NAS)
. Unlike previous approaches that apply search algorithms on a small, human-designed search space without considering hardware diversity, we propose HURRICANE that explores the automatic hardware-aware search over a much larger search space and a two-stage search algorithm, to efficiently generate tailored models for different types of hardware. Extensive experiments on ImageNet demonstrate that our algorithm outperforms state-of-the-art hardware-aware NAS methods under the same latency constraint on three types of hardware. Moreover, the discovered architectures achieve much lower latency and higher accuracy than current state-of-the-art efficient models. Remarkably, HURRICANE achieves a 76.67% top-1 accuracy on ImageNet with a inference latency of only 16.5 ms for DSP, which is a 3.47% higher accuracy and a 6.35x inference speedup than FBNet-iPhoneX, respectively. For VPU, we achieve a 0.53% higher top-1 accuracy than Proxyless-mobile with a 1.49x speedup. Even for well-studied mobile CPU, we achieve a 1.63% higher top-1 accuracy than FBNet-iPhoneX with a comparable inference latency. HURRICANE also reduces the training time by 30.4% compared to SPOS.
الأسئلة المقترحة
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
