4.4 KiB
Use Case and Architecture Sub-group Introduction
Project Facts
Project Creation Date: 2022/10/DD
Primary Contact: Qihui Zhao, zhaoqihui@chinamobile.com
Project Lead: Qihui Zhao, zhaoqihui@chinamobile.com
Committers:
- Weisen Pan, weisenpan@chinamobile.com
- Zhihua Fu, fuzhihua@h3c.com
- Xiaoqiao Geng, gengxiaoqiao@inspur.com
- Yong Chen, yong2.chen@nokia-sbell.com
- Jianchao Guo, TBD
- Yongfeng Nie, yongfeng.nie@huawei.com
- Weiguo Hu, hu.weiguo@99cloud.net
- Jie Niu, niujie@chinamobile.com
- Jian Xu, xujianwl@chinamobile.com
Mailing List: computing-force-network@lists.opendev.org
Meetings: No sub-group meeting time. Use bi-weekly meeting of CFN WG.
Repository: https://opendev.org/cfn/use-case-and-architecture
StoryBoard: https://storyboard.openstack.org/#!/project_group/computing-force-network
Open Bugs: N/A
Introduction
Use Case and Architecture is a sub group under CFN WG. It can give you an technical overview of CFN WG.
The goal of Use Case and Architecture sub group is to explore CFN related use cases, derive functional requirements based on use cases, give software recommendations to provide functions required by derived fucntional requriements, and maintain a system architecture and software landscape of CFN. Besides, this sub group is also a incubator of CFN subgroups and new feature. It will potentially provide code-level or mathematical or architecture-level implementations of features, software, systems and any for proper use cases before those deliveries be included in any other sub groups.
Documentation & Training
N/A
Release Planning & Release Notes
For the release of year 2024, Use Case and Architecture plans to accomplish the following task items:
- Landscape update
- As CFN key technologies and architecture have been discussed last year, an open source software landscape can be created for the WG to show what are included and how to integrate a tiny CFN. For this task, a landscape figure and explanation will be provided.
- Federated computing force network research
- A research of enhancing federated learning architecture through parameter optimization and neural architecture search in computing force network will be conducted. And a report will be provided.
- Analysis about CFN supporting AI use cases (optional)
- Taking LLM as the major use case, explore how CFN can support LLM being deployed and used, which may include CFN advantage analysis, specific AI use case and CFN capability discussion.
Previous Releases
For the release of year 2023, Use Case & Architecture sub group has accomplished the following task items:
- Sub-group introduction
- Link to this task: https://storyboard.openstack.org/#!/story/2010457
- Use case and architecture document version 1
- Link to the document: https://opendev.org/cfn/use-case-and-architecture/src/branch/master/CFN-Overview-&-Use-Case-Exploration-en.md
- Description: This doc discribes interpret the concept of CFN from user perspective, sort out the progress of CFN, analyze use cases and put forward suggestions on how to further promote maturity of CFN
- Reference algorithm implementation of AI O&M use case
- Link to the code: https://opendev.org/cfn/use-case-and-architecture/src/branch/master/ai_computing_force_scheduling
- Description: In CFN, computing tasks, network and computing resources are having tight collaboration. This implementation targets on creating models of task amount, network traffics and computing resource workloads, and using this models to predict the future status of the task, network and computing resources. With this prediction, CFN can adjust its network bandwidth, computing resource scaling in/out to better serve the task.
For the release of year 2023, Use Case & Architecture sub group will follow the unified timeline of CFN WG listed at https://opendev.org/cfn/cfn-overview/src/branch/master/releases/2023-cfn-release-timeline.md