|
Work item:
|
Y.AIoT-DA
|
|
Subject/title:
|
Framework of network resource management with data affinity in AIoT environments
|
|
Status:
|
Under study
|
|
Approval process:
|
AAP
|
|
Type of work item:
|
Recommendation
|
|
Version:
|
New
|
|
Equivalent number:
|
-
|
|
Timing:
|
2027-Q1 (Medium priority)
|
|
Liaison:
|
ITU-T SG13, SG17, SG21, ISO/IEC JTC 1/SC 41
|
|
Supporting members:
|
Korea (Rep. of), ETRI, Daejeon University
|
|
Summary:
|
A wide range of vertical domains, such as manufacturing, retail, energy, transportation, and healthcare, rely on data generated by IoT devices. Managing millions of these devices effectively is essential to support the rapid expansion of smart IoT services. Network management typically involves monitoring and controlling a network’s hardware, software, and traffic to ensure optimal performance and availability, while IoT network management interconnects massive resources from diverse devices and services. However, integrating IoT network resources into existing network management frameworks is highly complex, potentially limiting the scalability and resilience of IoT networks. The convergence of artificial intelligence (AI) and IoT addresses this challenge by enabling intelligent decision-making and automating complex processes through data and analytics.
To reduces the complexity of the management process, a grouping approach called an “affinity group” streamlines management operations by logically organizing resources into families based on their function, location, or other relevant criteria. The concept of data affinity, originally introduced in cloud computing, involves placing two or more resources in close proximity to enhance bandwidth and improve communication between them. The data affinity in grouping seeks to enable intelligent management and provide flexible control over network resources. As a result, network resources within the same affinity group share similar attributes and semantic properties. Moreover, artificial intelligence (AI) in IoT further advances affinity grouping by uncovering hidden relationships between network resources. The relational intelligence derived from the data affinity benefits management of network resources (e.g., software, hardware, etc.) by enhancing efficiency and scalability.
This draft recommendation aims to establish a framework for network resource management with data affinity in AIoT environments. As a key enabling technology, affinity grouping facilitates intelligent management and flexible control over network resources. Ultimately, the relational intelligence derived from data affinity streamlines decision-making processes and enhances the scalability of IoT services.
|
|
Comment:
|
-
|
|
Reference(s):
|
|
|
Historic references:
|
|
Contact(s):
|
|
| ITU-T A.5 justification(s): |
|
|
|
|
First registration in the WP:
2025-01-29 16:03:56
|
|
Last update:
2026-06-16 16:07:39
|
|
