| Internet-Draft | Adaptive Traffic Data Collection | September 2022 |
| He, et al. | Expires 23 March 2023 | [Page] |
IP carrier network needs to provide real-time traffic visibility to help network operators quickly and accurately locate network congestion and packet loss, and make timely path adjustment for deterministic services in order to avoid congestion. It is essential to explore the adaptive traffic data collection mechanism so as to capture real-time network state at minimum resource consumption.¶
This document summarizes the problems currently faced by network operators when attempting to provide timely traffic data collection to satisfy various scenarios that require real-time network state and traffic visibility, and aggregates the requirements for adaptive traffic collecting mechanism from a variety of deployment scenarios.¶
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 23 March 2023.¶
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.¶
With the advent of cloud computing, big data and Artificial Intelligence (AI) , as well as the large-scale deployment of 5G mobile communication technology, a large number of ultra Reliable & Low Latency Communication (uRLLC) services such as Augmented Reality (AR)/Virtual Reality (VR), Industrial Internet and Computing Power Network (CPN) have emerged, which puts forward higher requirements for service quality of IP carrier networks. IP carrier networks need to provide real-time traffic visibility to help network operators quickly and accurately locate network congestion and packet loss, and make timely path adjustment for the services of deterministic delay in order to avoid the congested nodes and links. For such business scenarios, the network needs to provide traffic sampling interval of sub seconds or even milliseconds level so as to gain real-time network state.¶
For decades, SNMP [RFC3416] and the Management Information Bases (MIBs) have been widely deployed and the de facto choice for many monitoring solutions, especially in collecting interface traffic. Arguably the biggest shortcoming of SNMP for those applications concerns the need to rely on periodic polling, because it introduces an additional load on the network and devices, and it is brittle if polling cycles are missed. Therefore, SNMP has no capability to realize real-time traffic sampling at sub seconds or even milliseconds intervals. Telemetry, as a revolutionary data acquisition technique, based on pull mechanism that is able to deliver object changes as they happen, overcomes the limitations of SNMP such as "low sampling rate, inefficiency and more processing resources". Nevertheless, for the sake of capturing real- time network state, persistent sampling of interface traffic at milliseconds intervals will generate a considerable amount of data which may claim too much transport bandwidth and overload the servers for data collection, storage and analysis. Increasing the data handling capacity is technically feasible but expensive, and difficult to achieve large-scale deployment in operator's networks. It is essential to explore the adaptive traffic data collection mechanism so as to capture real-time network state at minimum resource consumption.¶
This document summarizes the problems currently faced by network operators when attempting to provide timely traffic data collection to satisfy the aforementioned new services and applications that require real-time network state and traffic visibility. Also, this document aggregates the requirements for adaptive traffic collection mechanism from a variety of deployment scenarios.¶
Artificial Intelligence¶
Augmented Reality¶
Virtual Reality¶
Computing Power Network¶
Google Network Management Interface¶
IP Radio Access Network¶
Deterministic Networking¶
Quality of Experience¶
Service Level Agreement¶
ultra Reliable & Low Latency Communication¶
Network Management System¶
Internet Data Center¶
Simple Network Management Protocol¶
Management Information Base¶
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.¶
The following terms are defined in this document:¶
Allow servers automatically switch to different telemetry sampling period to collect traffic data according to the threshold change.¶
As is well known ,IP network, based on statistical multiplexing model, is of traffic burst characteristic. In order to avoid congestion, network operators have been keeping network utilization at a rather low level. For a long time, operators have obtained traffic visibility from the Network Management System (NMS), and satisfied with 30~40% bandwidth utilization from traffic statistics curves. In spite of such a low link usage, many complaints have still been received about poor Quality of Experience (QoE) in delivering applications with the sensitivity of delay and packet loss. The fundamental cause lies in that the observed average network traffic at every sampling cycle masks the characteristic of traffic burst, given that SNMP is widely employed in operator's networks to collect interface traffic at 5 minutes intervals. Because of low sampling rate, SNMP has no capability to capture traffic burst characteristic.¶
A large quantity of laboratory data and network operational data indicate that a microburst phenomenon occurs frequently in operator's carrier networks, such as IP based Radio Access Network (IP RAN), IP metropolitan network, IP backbone network and Internet Data Center (IDC). The typical duration of such a microburst is tens to hundreds of milliseconds, easy to cause instantaneous congestion of interface output queue. Network congestion amplifies queuing delay and jitter, and in severe cases, it may even cause packet loss. Thus, the congestion caused by microburst is not beneficial to the deterministic-delay applications. The congestion problem is a major challenge for IP networks, and the congestion caused by microburst is difficult to eliminate, but must be avoided.¶
Although the mechanism of microburst is not very distinct, it does not hinder us to detect it. Fortunately, Telemetry (e.g., YANG PUSH [RFC8639] [RFC8641], gNMI [gNMI]) has the capability to collect interface traffic at a higher frequency, i.e., milliseconds interval. So, by means of telemetry technique, we can capture the complete aspects of a microburst traffic. However, it is impractical to gain the real-time traffic visibility at the cost of persistent sampling at millisecond intervals. For example, in order to capture a microburst traffic of interface, at least 10-millisecond sampling cycle is necessary, and as a result, the required resources for data storage and analysis will increase by 30000 times, compared with the today's widely employed 5-minute sampling cycle based on SNMP.¶
It is essential to investigate the adaptive traffic data collection mechanism so as to capture real-time network state at minimum resource consumption. Generally speaking, under normal non-congested network conditions, which happen at the time of 95% above, minutes-level sampling cycle is enough because of almost invariable forwarding delay and less jitter of interface. However, when detecting a congestion state or congestion trend, sampling period must be timely tuned to milliseconds to capture a microburst traffic of interface. A congestion state or congestion trend of interface manifests itself in the form of packet loss due to queue overflow, queue depth beyond the threshold or too high link utilization, which can be defined as Event-triggered data. Such event data can be actively pushed through subscription or passively polled through query. Although the microburst phenomenon occurs frequently, it is transient and a real-time detection tool is preferable to pinpoint it timely. The traditional method of using CPU on main control board through query is processing resources consuming, the network device must possess built-in hardware designed specially to monitor it.¶
In order to reduce the excessive consumption of resources caused by milliseconds-level collection of the single data, batch data such as hundreds of sampled traffic data from an interface can be packaged as a telemetry packet and is sent to the collector. The timestamp is required for every sampled traffic data for the convenience of the collector visualizing the interface traffic trend in the form of curve. And the collector must make traffic visualization in real-time manner so that the operators can observe it immediately.¶
This section presents several typical scenarios which require adaptive traffic data collection to gain real-time network state and traffic visibility at minimum resource consumption.¶
Interface traffic data collection is one of the most important functions for NMS. Today, more and more applications are of latency-sensitive and loss-sensitive characteristic, and the real-time traffic visibility can help operators better understand network performance so as to achieve SLA guarantees. On the other hand, obtaining the holistic and genuine characteristic of interface traffic is also a basic requirement for the statistical multiplexing model of IP network, which is of great significance for traffic prediction, network planning, network capacity expansion, network optimization, etc. For example, a higher long-term average utilization prompts need of capacity expansion; a higher ratio between the peak and the average, as well as frequent microbursts detected, implies a intense traffic burst characteristic, suggesting the timely path adjustment for those key traffic flows of deterministic delay. However, the traditional NMS based on SNMP has no capability to depict genuine characteristic of interface traffic, and interface traffic data collection based on telemetry techniques is preferable.¶
It is essential to exploit the adaptive traffic data collection techniques to depict multi-dimensional real-time portrait of interface traffic characteristic at minimum resource consumption. That is to say, in normal non-congested network conditions, which happen at the time of 95% above, minutes-level sampling cycle is enough as it is. But, while detecting a congestion state or congestion trend, sampling cycle must be timely tuned to milliseconds to capture a microburst traffic of interface. Such an adaptive traffic data collection technique can not only reflect the coarse-grained interface traffic characteristics, but also capture the congestion state of interface with finer time granularity. Because the traffic data collection with very high rate is seldom (i.e., only triggered by the detected microbursts), we can depict multi-dimensional real-time portrait of interface traffic characteristic at minimum resource consumption. Because of the lower cost, it can be deployed on large-scale in operator's networks.¶
Microburst traffic, as an instantaneous congestion phenomenon occurring frequently in IP carrier network, will cause critical delay jitter and even packet loss, which will seriously affect the QoE of latency-sensitive and loss-sensitive applications. The ability of detecting microburst traffic of interface will help network operators quickly and accurately locate network congestion and packet loss, and make timely path adjustment for deterministic-delay services in order to avoid the congested nodes and links. In order to have a comprehensive understanding of microburst, we must timely collect interface traffic as soon as it occurs. For example, how often does it occur? and what duration does it last? only event data representing a microburst such as packet loss and queue length beyond threshold are not enough to describe its characteristic.¶
Because the typical duration of such a microburst is generally tens to hundreds of milliseconds, at least 10-millisecond sampling cycle is necessary. Although the microburst phenomenon occurs frequently, it takes very little time of 24 hours a day. It is not a good approach to observe it through persistent millisecond sampling period. Preferably, we can capture it as soon as a microburst occurs to ensure important diagnose data will not be missed. Because it is transient, and an on-line detection tool based on the dedicated hardware is required to pinpoint it timely. Triggered by the events such as packet loss, queue depth beyond the threshold which is detected timely, sampling period must be timely tuned to milliseconds to capture a microburst traffic of interface. In a word, it is of practical significance to explore the microburst detection technique aiming at minimizing resource consumption.¶
Network congestion will rapidly increase queuing delay and jitter, and may even give rise to packet loss, which will seriously affect the QoE of delay-sensitive and packet loss-sensitive applications. The goal of network optimization is to reduce the occurrence of network congestion as much as possible.¶
It is a complicated problem for network operators to accurately predict the trend of network congestion and make network adjustment in advance. The real-time traffic visibility based on the adaptive traffic data collection techniques can accurately predict the long- term congestion, and quickly capture the instantaneous congestion (i.e., microburst) of interface. By means of the real-time traffic visibility, the automatic optimization tool (e.g., AI) can make timely path adjustment for key traffic flows. For example, based on the real-time traffic visibility and microburst events (e.g., packet loss, queue depth) collected, the controller can accurately predict the congestion trend of interface and make timely traffic redirection to the non-congested interface for deterministic delay applications.¶
On-path telemetry (e.g., IOAM [RFC9197]) is useful for application-aware networking operations. For example, it is critical for the operators who offer high-bandwidth, latency and loss-sensitive services such as video streaming and online gaming to closely monitor the relevant flows in real-time as the basis for any further optimizations. Applying on- path telemetry on all packets of the selected flows is resource consuming. A sampling rate should be set for these flows and only enable telemetry on the sampled packets. However, a too high rate would exhaust the network resource and even cause packet drops; an overly low rate, on the contrary, would result in the loss of information and inaccuracy of measurements.¶
An adaptive approach can be used based on the network conditions to dynamically adjust the sampling rate. In normal network state, a low sampling rate is enough to reflect network performance (i.e., almost invariable forwarding delay and less jitter of interface) ; But, in the case of network congestion, the controller is aware of it from the real- time traffic visibility and events data collected (e.g., packet loss, queue depth), and timely adjust the packet sampling rate at very high level. Even all packets of the selected flows are applicable to be sampled so as to acquire actual measurement data such as latency, jitter and packet loss.¶
Similarly, such an adaptive approach can applicable to the traditional active measurement methods (e.g., a Two-Way Active Measurement Protocol (TWAMP)[RFC5357]), so as to improve measurement accuracy at minimal resource consumption. In the case of normal non-congested conditions, the probing packets are send at longer intervals, But, in case of network congestion caused by microburst, the controller is aware of it from the real- time traffic visibility, and change the probing packets to the shorter intervals timely, which can capture the microburst traffic and therefore get real measurements of congestion state.¶
This document does not include an IANA request.¶
This document provides an adaptive telemetry mechanism to minimize the resource consumption. The increased complexity of network telemetry may give rise to some security concerns. For example, persistent traffic collection at very high rate (e.g., at milliseconds intervals) induced by misconfiguration or spurious triggering might exhaust resources of network device as well as the collector; Also, an inappropriate threshold setting which trigger high sampling rate should be avoided. Therefore, access control for enabling and disabling adaptive telemetry is required , also, rate control for collecting telemetry data is recommended so as to avoid degradation of network performance.¶
On the other hand, for security considerations of telemetry management interface such as NETCONF or gNMI, it must provide authentication, data integrity,confidentiality, and replay protection. The lowest NETCONF layer is the secure transport layer, and the mandatory-to-implement secure transport is Secure Shell (SSH) [RFC6242]. The lowest gNMI layer is HTTPS, and the mandatory-to-implement secure transport is TLS [RFC5246]. And further study of the security issues will be required.¶