|
Work item:
|
Y.AINLB-req-fra
|
|
Subject/title:
|
Requirements and framework for AI/ML-based network load balancing in future networks
|
|
Status:
|
[Carried to next study period]
|
|
Approval process:
|
AAP
|
|
Type of work item:
|
Recommendation
|
|
Version:
|
New
|
|
Equivalent number:
|
-
|
|
Timing:
|
-
|
|
Liaison:
|
ITU-T SG2, 3GPP, ETSI, GSMA, IETF
|
|
Supporting members:
|
China Telecom, Huawei Technologies Düsseldorf GmbH, China Unicom, ZTE
|
|
Summary:
|
New applications and services, e.g., cloud-based AR/VR, and cross-node distributed machine learning, have led to a significant surge in traffic volume. These applications and services bring distinct traffic patterns characterized by large individual flow sizes. Additionally, the increased number of service flows and other traffic within the network makes the overall traffic load even more complex. Efficient network load balancing is crucial under this increased and complicated traffic load. Imbalances in load distribution can result in network congestion, impacting various services and lowering transmission efficiency.
This Recommendation specifies the requirements, functional framework, and procedures for AI/ML-based network load balancing. The study aims to specify efficient load balancing through real-time network status identification and dynamic traffic scheduling using AI/ML, thereby improving overall network throughput and performance.
|
|
Comment:
|
-
|
|
Reference(s):
|
|
|
Historic references:
|
|
Contact(s):
|
|
| ITU-T A.5 justification(s): |
|
|
|
|
First registration in the WP:
2024-08-08 10:34:20
|
|
Last update:
2024-09-19 10:41:45
|
