Hi Tony, Thank you for your comment. I fully agree with your observation that it is crucial to reduce the risk of oscillation when propagating real-time topology and link congestion information across the network. Since link congestion information is significantly dynamic, it's essential to use the threshold of available link capacity variation to trigger and suppress an update. At least, link capacity information that is relatively stable can be used to achieve global load-balancing, particularly when multiple physical links are deployed between peers.
Some network chip vendors have adopted their proprietary signals to propagate congestion information for the purpose of global load-balancing. The draft just proposes an alternative approach based on open standards which could be used across different network chips. Best regards, Xiaohu 发件人: Tony Li <tony1ath...@gmail.com> 代表 Tony Li <tony...@tony.li> 日期: 星期五, 2023年11月24日 00:36 收件人: xuxiaohu_i...@hotmail.com <xuxiaohu_i...@hotmail.com> 抄送: lsr@ietf.org <lsr@ietf.org> 主题: Re: [Lsr] New Version Notification for draft-xu-lsr-fare-00.txt Hi Xiaohu, One way of achieving this would be to use the Unreserved Bandwidth TLV (https://datatracker.ietf.org/doc/html/rfc5305#autoid-10) to report the unused bandwidth on a link. Then, you would have to explain how this does not become an oscillator. I’m not optimistic. Regards, Tony On Nov 23, 2023, at 8:27 AM, xuxiaohu_i...@hotmail.com wrote: Hi all, Any comments or suggestions are welcome. Best regards, Xiaohu 发件人: internet-dra...@ietf.org <internet-dra...@ietf.org> 日期: 星期五, 2023年11月24日 00:13 收件人: Xiaohu Xu <xuxiaohu_i...@hotmail.com> 主题: New Version Notification for draft-xu-lsr-fare-00.txt A new version of Internet-Draft draft-xu-lsr-fare-00.txt has been successfully submitted by Xiaohu Xu and posted to the IETF repository. Name: draft-xu-lsr-fare Revision: 00 Title: Fully Adaptive Routing Ethernet Date: 2023-11-22 Group: Individual Submission Pages: 7 URL: https://www.ietf.org/archive/id/draft-xu-lsr-fare-00.txt Status: https://datatracker.ietf.org/doc/draft-xu-lsr-fare/ HTMLized: https://datatracker.ietf.org/doc/html/draft-xu-lsr-fare Abstract: Large language models (LLMs) like ChatGPT have become increasingly popular in recent years due to their impressive performance in various natural language processing tasks. These models are built by training deep neural networks on massive amounts of text data, often consisting of billions or even trillions of parameters. However, the training process for these models can be extremely resource- intensive, requiring the deployment of thousands or even tens of thousands of GPUs in a single AI training cluster. Therefore, three- stage or even five-stage CLOS networks are commonly adopted for AI networks. The non-blocking nature of the network become increasingly critical for large-scale AI models. Therefore, adaptive routing is necessary to dynamically load balance traffic to the same destination over multiple ECMP paths, based on network capacity and even congestion information along those paths. The IETF Secretariat _______________________________________________ Lsr mailing list Lsr@ietf.org https://www.ietf.org/mailman/listinfo/lsr
_______________________________________________ Lsr mailing list Lsr@ietf.org https://www.ietf.org/mailman/listinfo/lsr
