COINRG C. Li, H.Yang, Z. Sun Internet-Draft Beijing University of Posts and Telecommunications Intended status: Standards Track S. Liu Expires: 10 July 2023 China Mobile Research Istitute H. Zheng Huawei Technologies 10 January 2023 Distributed Learning Architecture based on Edge-cloud Collaboration draft-li-coinrg-distributed-learning-architecture-01 Abstract This document describes the distributed learning architecture based on edge-cloud collaboration. Status of This Memo This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet- Drafts is at https://datatracker.ietf.org/drafts/current/. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress." This Internet-Draft will expire on 10 July 2023. Copyright Notice Copyright (c) 2022 IETF Trust and the persons identified as the document authors. All rights reserved. This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/ license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Revised BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Revised BSD License. Li, et al. Expires 10 July 2023 [Page 1] Internet-Draft Distributed Learning Architecture January 2023 Table of Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 1.1. Requirements Language . . . . . . . . . . . . . . . . . . 3 2. Scenarios . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2.1. Federated Learning . . . . . . . . . . . . . . . . . . . 3 2.2. Model Parallelism-Based Distributed Training . . . . . . 4 3. Problem Statement . . . . . . . . . . . . . . . . . . . . . . 4 4. Distributed Learning Architecture based on Edge-cloud Collaboration. . . . . . . . . . . . .. . . . . . . . . . . . . 5 4.1. Model Splitting . . . . . . . . . . . . . . . . . . . . . 5 4.2. Distributed Learning Architecture based on Edge-cloud Collaboration . . . . . . . . . . . . . . . . . . . . . . . . . 5 5. Manageability Considerations . . . . . . . . . . . . . . . . 7 6. Security Considerations . . . . . . . . . . . . . . . . . . . 7 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 7 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 7 Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . 8 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 8 1. Introduction The rapid growth of Internet of Things (IoT) and social networking applications has led to exponential growth in the data generated at the edge of the network. The ability of a single edge node to process data cannot meet the needs of IoT services. Edge-cloud collaboration technology emerged as the times require, offloading some computing tasks at the edge to the cloud. Service latency includes edge-side computing latency and service transmission latency, which is crucial to model quality in distributed training based on edge-cloud collaboration, because it affects the synchronization of training. How to ensure these two delays has become a key factor in improving the quality of the model. The distributed learning architecture based on edge-cloud collaboration has become a solution to the above problems. The training tasks are flexibly deployed to edge devices and cloud devices through model parallelism, and deterministic network technology is used to ensure uniform edge training delay and model transmission delay, and then distributed training technology is used to generate a unified model. Li, et al. Expires 10 July 2023 [Page 2] Internet-Draft Distributed Learning Architecture January 2023 1.1. Requirements Language The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all capitals, as shown here. 2. Scenarios In recent years, with the combination of edge computing and AI, edge AI has gradually become a new means of intelligence transformation due to its small traffic footprint, low latency, and privacy features. Distributed edge model training can be the main means to achieve edge intelligence. 2.1. Intelligent Transportation With the development of urban traffic intelligence, the number of various terminals increases, and the demand for real-time processing of massive amounts of information increases significantly. For example, traffic surveillance cameras, HD cameras at a single intersection generate tens of G's of video files every day, and if it is a street, a region, or even a city, the amount of data generated is immense, and the content of these videos that are truly effective and need to be captured for illegal behavior accounts for very little. Edge AI computers can perform intelligent processing in the field, analyze violations directly locally, and filter valuable content for upload, greatly reducing the waste of bandwidth and storage generated by invalid content. However, the effective data collected by a single computer is very limited, and the computers themselves have very limited computing power, making it difficult to train a high-precision AI model. The above challenges can be effectively solved by the distributed collaborative training method in this paper. Li, et al. Expires 10 July 2023 [Page 3] Internet-Draft Distributed Learning Architecture January 2023 2.2. Smart Factory In the field of industrial manufacturing, edge AI will play an increasingly important role in the development of smart factories. Driven by the Industry 4.0 model, smart factories will apply advanced robotics and machine learning technologies to software services and industrial IoT to improve production capacity and maximize productivity. Edge AI uses a variety of sensors to control and manage commands, significantly improving control efficiency and reducing errors. Edge AI computers can independently and autonomously respond to inputs within milliseconds, either making adjustments to fix the problem or immediately stopping the production line to prevent a serious safety incident. However, the limited computing power of in-plant edge computers makes it difficult to train models with high accuracy. The problem can be effectively solved by federated learning and collaborative training. 3. Problem Statement The computing power of edge nodes is small and cannot meet the model training in the case of a large amount of data. Therefore, distributed training based on edge-cloud computing power coordination has become an important means to realize edge intelligence. In order to obtain good training results, distributed training based on edge-cloud collaboration requires the deterministic performance of the underlying optical network. The synchronization of distributed training is achieved through deterministic performance. At this time, it is necessary to synchronize the edge training delay and model transmission delay. These require the support of various quality factors, such as computing resources, end-to-end delay, delay jitter, bandwidth. The above factors can be achieved by deterministic optical networks. Li, et al. Expires 10 July 2023 [Page 4] Internet-Draft Distributed Learning Architecture January 2023 4. Distributed Learning Architecture based on Edge-cloud Collaboration At present, the common method is to realize the training of distributed models by combining model splitting and distributed training. 4.1. Model Splitting Since each layer of an artificial intelligence model has independent inputs and outputs, a model can be split into multiple sub-models for independent training, where the training layer that links the sub- models is called a segmentation layer. This method provides the realization basis for edge-cloud collaborative training. In order to maintain the synchronization in the data parallel process, the training time of all edge nodes in this paper needs to be consistent. Before the model is divided, the computing resources required by each layer are first calculated, and the model is splited according to the remaining situation of the current computing resources. The model splitting in this document is dynamic, that is, the splitting scheme of the model may be different for each round of training. 4.2.Distributed Learning Architecture based on Edge-cloud Collaboration Edge devices provide services to nearby users, and collect data generated in the process of providing services in real time to form edge data sets. After the edge collects enough edge data, it sends a model training request to the cloud node. After the cloud node receives all training requests from the edge device, it prepares for model training, which is divided into data standardization and model determination. Model determination: The cloud node determines the model architecture according to the training task and sends it to all edge devices. In order to reduce the amount of computation in the training process, the dataset needs to be standardized before training. Common methods include normalization, log transformation, and regularization. The data standardization method is determined by the cloud node, and the standardized algorithm is sent to the edge device, and the edge device processes the edge data set according to the standardized algorithm. Li, et al. Expires 10 July 2023 [Page 5] Internet-Draft Distributed Learning Architecture January 2023 After the preparations are completed, enter the model training phase. In order to ensure the quality of model training, it is necessary to ensure the consistency of training delay in all edge devices and the consistency of model transmission delay. Training delay and transmission delay are set by cloud nodes based on historical experience. At present,the computing power network calculation can calculate the training time of the training task. Therefore, in terms of the training delay, combining the model splitting and the computing power network can calculate the training time of each layer of the model, and then calculate the edge device according to the training delay. The number of layers to train. At the same time, it is also possible to determine the size of the data volume of the segmentation layer, and then reserve bandwidth for model transmission in advance based on the determined network technology. The edge device finishes training the pre-training model, and after the training is completed, sends the segmentation layer of the pre-training model to the cloud node. After receiving the segmentation layer of the model, the cloud node completes the subsequent training of the model, and then updates the model weights according to the back-propagation algorithm. So far, the edge device and the cloud node have completed a round of model training. After every 5 rounds of training, all edge devices generate a global model through distributed learning, and edge devices continue to train according to the local model according to the global model until the model accuracy meets the requirements. Li, et al. Expires 10 July 2023 [Page 6] Internet-Draft Distributed Learning Architecture January 2023 5. Manageability Considerations TBD 6. Security Considerations TBD 7. IANA Considerations This document requires no IANA actions. 8. References TBD Li, et al. Expires 10 July 2023 [Page 7] Internet-Draft Distributed Learning Architecture January 2023 Acknowledgments TBD Authors' Addresses Chao Li Beijing University of Posts and Telecommunications Email: lc96@bupt.edu.cn Hui Yang Beijing University of Posts and Telecommunications Email: yanghui@bupt.edu.cn Zhengjie Sun Beijing University of Posts and Telecommunications Email: sunzhengjie@bupt.edu.cn Sheng Liu China Mobile Email: liushengwl@chinamobile.com Haomian Zheng Huawei Technologies Email: zhenghaomian@huawei.com Li, et al. Expires 10 July 2023 [Page 8]