The Changsha station of the cloud native community was established in June 2021. The community is committed to promoting cloud native technology and building a developer ecosystem. After the establishment of the community, two offline meetings and brainstorming sessions were held. Cloud native technology enthusiasts are welcome to join. Welcome to the native community Changsha Station (Changsha students can add WeChat zb016720 and apply to join the group).
After waiting and preparing for a long time, the Cloud Native Community Meetup Changsha Station is finally about to start.
event information
•Event Date: December 3rd
• Event time: 14:00 - 17:00
• Venue: Headquarters Building, CLP Software Park, No. 39 Jianshan Road, Yuelu District, Changsha City
• Event scale: 50 - 80 people, offline
• Sponsor: Cloud Native Community
• Organizer: Yunyuan Community Changsha Station
•Venue & Tea Break Sponsorship: CLP Software Park, GitLab
• Cooperative community: openKylin, Tengyuan Club
issue
Sign up
Scan the QR code below to start registering!
Event Schedule
13:30 - 13:50 Check-in
13:50 - 14:00 Moderator opening
Topic 1
14:00 - 14:40: How to improve R&D efficiency while ensuring safety?
Lecturer : Ma Jinghe
self introduction:
Jihu (GitLab) DevOps technical evangelist, LFAPAC open source evangelist, CDF ambassador. Focus on cloud native and DevSecOps.
Topic outline:
Efficiency, quality, safety and compliance are three crucial aspects in the software development process. It is extremely challenging to realize the three aspects simultaneously in the software development process. It is not only related to the skills of the team members, but more importantly, it is related to the team It is related to the research and development process, the thinking awareness of team members, and team culture. The confusion of the process and the inconsistency of awareness make improving efficiency, ensuring quality, and ensuring safety and compliance a slogan that is mere formality and difficult to implement. Jihu GitLab workflow forms a workflow that can standardize and standardize the R&D process by "stringing" project requirements management, branch development, Code Review, Approve Rule, CI/CD, Security, etc., through configuration and process. It is possible to improve efficiency, ensure quality, and ensure safety and compliance in the software development process.
Audience benefits:
1. How Jihu GitLab workflow standardizes the R&D process
2. How Code Review + Approve Rule improves code quality
3. Let’s see how a few lines of code can solve DevSecOps
Topic 2
14:40 - 15:20: Cluster management based on ClusterAPI
Lecturer : Guo Zhiwei
self introduction:
A member of the China Mobile Jinke cloud platform team, has been engaged in the cloud computing industry for many years. Currently focused on cloud native infrastructure.
Topic outline:
Today, with platform engineering and SRE widely implemented, infrastructure as code provides a complete methodology and practical basis for efficient operations. For the infrastructure team, how to use GitOps to declaratively manage the life cycle of k8s clusters, how to quickly deploy, expand and shrink on demand, and roll upgrade k8s clusters. The k8s SIG gave a reference - ClusterAPI.
Audience benefits:
The lecturer will introduce the basic design of ClusterAPI, project ecology, and application scenarios in infrastructure construction.
Topic 3
15:35 - 16:15: Operating and maintaining agile organizations
Lecturer : Tang Botao
self introduction:
Operation and maintenance engineer, engaged in solutions, demand analysis, project implementation and delivery, application operation and maintenance, security operation and maintenance, capacity planning, etc.
Topic outline:
In the era of popular cloud services, there are endless rumors about eliminating operation and maintenance. How can the operation and maintenance person in charge of operation and maintenance adapt to the VUCA environment, and how to lead the team to upgrade and transform in an uncertain environment?
Audience benefits:
This live broadcast will share the lecturer's thinking on this issue, and build an agile organization integrating research and operation as a solution.
Topic 4
16:15 - 16:55: Sharing of the latest cloud desktop open source achievements: Analysis of virtio-gpu video acceleration framework
Lecturer : Jiang Feng
self introduction:
openKylin Virtualization SIG Maintainer, senior R&D engineer of the virtualization team of Kylin Software Co., Ltd.
Topic outline:
In virtualization application scenarios such as cloud desktops, commonly used virtual graphics solutions include QXL, Virtio GPU, etc. However, QXL is a purely software-simulated virtual graphics card, which relies entirely on the CPU for image rendering and video processing, resulting in weaker performance. In contrast, the Virtio GPU can achieve 3D hardware acceleration with the help of the virglrenderer component to quickly increase graphics performance. But in terms of video processing, Virtio GPU, like QXL, still uses software codec, which has obvious performance shortcomings. Therefore, we designed and developed a video acceleration mechanism for Virtio GPU, and related patches have been adopted by the upstream community. The mechanism has realized the hardware decoding function of H.264 and H.265 under the Virtio GPU, and greatly reduced the CPU usage and greatly improved the user experience. This sharing will introduce the mechanism in detail. Principles and practical effects.
Audience benefits:
• Understand the development status of Virtio GPU technology in cloud desktops, and the latest hardware video acceleration framework based on Virtio GPU.
•By using this framework, the application program in the virtual machine can use the graphics card on the physical machine to perform hardware video encoding and decoding,
•Thereby greatly improving the codec performance and improving the video playback and editing experience of cloud desktop users.