OPSA F. Zhao Internet-Draft Huawei Intended status: Informational 21 August 2023 Expires: 22 February 2024 Near Data Processing for Telemetry draft-zhao-opsawg-ndp-telemetry-00 Abstract As the scale of IP networks and the importance of services increase continuously, the data collected by telemetry is increasing exponentially. To support the network simulation, traffic optimization and risk detection of the network, Digital Twin technology on data communication networks is mentioned increasingly. Real-time synchronization between controllers and devices is a critical feature of Digital Twin Network, which require much more volume telemetry data between controllers and devices. All of these bring more pressure on the network bandwidth, device CPU and controller CPU. This document proposes a method to optimize the problem. Requirements Language The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in RFC 2119 [RFC2119]. 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 22 February 2024. Zhao Expires 22 February 2024 [Page 1] Internet-Draft ndp-telemetry August 2023 Copyright Notice Copyright (c) 2023 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. Table of Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 2. Traditional telemetry is extremely applied . . . . . . . . . 3 3. Near data processing and distributed computing method . . . . 3 4. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 4 6. Security Considerations . . . . . . . . . . . . . . . . . . . 4 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 4 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 4 8.1. Normative References . . . . . . . . . . . . . . . . . . 4 8.2. Informative References . . . . . . . . . . . . . . . . . 4 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 4 1. Introduction To implement more accurate, reliable, and timely network control, more and more Telemetry data is collected from network devices to controllers. In the future, the proposed and application of digital twin network has higher requirements for the real-time consistency between network devices and controllers. Meanwhile, with the widely used heterogeneous computing hardware and AI technologies, network devices will have more and more computing resources. We can consider migrating the analysis, data association, and even closed-loop decision-making processes implemented by controllers to the NE side to make full use of distributed computing resources. Similar frameworks have been maturely applied in the automotive self- driving field. In automotive self-driving field, BEV technologies are used to merge multi-dimensional data(such as camera, radar and position) and make self-decision at the car side. Zhao Expires 22 February 2024 [Page 2] Internet-Draft ndp-telemetry August 2023 2. Traditional telemetry is extremely applied Network devices collect information from multiple dimensions, including flow information, configuration, events, alarms, logs, dynamic topology and routes, and device status (including CPU, memory, and hardware health). To implement accurate network management, a large amount of data is collected to the controller for analysis and processing. This brings great challenges to the controller in terms of bandwidth resources, CPU resources. In order to try to optimize this problem, many methods have been proposed, such as Sketch, variable frequency sampling, data compression, and so on. However, these techniques are difficult to achieve a good balance between the integrity and the amount of data collected. 3. Near data processing and distributed computing method We consider a near-data and distributed computing scheme that makes full use of the computing power of network devices to realize real- time sensing and control of network status. It is also the foundation for the realization of the future digital twin network. Here's a framework for addressing the problem: +--------------+ | | | Controller | | | +--------------+ ^ | | result | +--------------------------+--------------+ | +----------------+ | | Flow Data ---->| | | | | | | | Event ---->| NDP Operation | | | | | | | Configuration ---->| | | | +----------------+ | +-----------------------------------------+ Device side: The CPU and AI computing capabilities of devices are used to converge and process original device data. For example, the flow data, events, and network topology can be comprehensively analyzed to determine the root cause of route flapping. Then only the results are sent to the controller. In this way, the original Zhao Expires 22 February 2024 [Page 3] Internet-Draft ndp-telemetry August 2023 data and result data can be sent. Of course, how to define and design a new result-based Telemetry model is a matter that needs further consideration in the future. 4. Summary This draft describes a lightweight and efficient method to synchronize data between devices and controller in really time. Going forward, how to describe the computing model on the device and the interface shall be studied to extend the draft. 5. IANA Considerations This document makes no request of IANA. Note to RFC Editor: this section may be removed on publication as an RFC. 6. Security Considerations TBD 7. Acknowledgements TBD 8. References 8.1. Normative References [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, DOI 10.17487/RFC2119, March 1997, . 8.2. Informative References [InfRef] "", 2004. Author's Address Fenghua Zhao Huawei Nanjing China Phone: +86-15805198370 Email: zhaofenghua@huawei.com Zhao Expires 22 February 2024 [Page 4]