KubeCon + CloudNativeCon + Open Source Summit + AI

In-person
21-23 August, 2024
Learn More and Register to Attend

The Sched app allows you to build your schedule but is not a substitute for your event registration. You must be registered for KubeCon + CloudNativeCon + Open Source Summit + AI_Dev China 2024 to participate in the sessions. If you have not registered but would like to join us, please go to the event registration page to purchase a registration.

Please note: This schedule is automatically displayed in Hong Kong Standard Time (UTC +8). To see the schedule in your preferred timezone, please select from the drop-down menu to the right, above "Filter by Date." The schedule is subject to change and session seating is available on a first-come, first-served basis.

亲临现场

2024年8月21-23日

了解更多并注册参加

Sched应用程序允许您创建自己的日程安排，但不能替代您的活动注册。您必须注册参加KubeCon + CloudNativeCon + Open Source Summit + AI_Dev China 2024，才能参加会议。如果您尚未注册但希望加入我们，请访问活动注册页面购买注册。

请注意：本日程自动显示为香港标准时间（UTC +8）。要查看您偏好的时区的日程，请从右侧“按日期筛选”上方的下拉菜单中选择。日程可能会有变动，会议席位先到先得。

11:00 HKT

How to Increase the Throughput of Kubernetes Scheduler by Tens of Times | 如何将Kubernetes调度器的吞吐量提高数十倍 - Yuquan Ren & Bing Li, ByteDance

Wednesday August 21, 2024 11:00 - 11:35 HKT

Level 1 | Hung Hom Room 2

Currently, various Kubernetes-based task schedulers popular in the community have limited performance capabilities, which restricts the cluster scale they can handle. Due to the limitation of cluster scale, it is difficult to improve resource utilization through large-scale colocation, and more clusters also bring greater operational burdens. 1. Due to the bottleneck of the scheduler and related components, the maximum cluster scale cannot exceed 5k nodes; 2. In clusters with more than 5k Nodes, scheduling throughput cannot exceed 100 Pods/s. Godel Scheduler is a distributed high-performance scheduler based on Kubernetes, and it is now open-sourced. In this talk, we will go deep into the performance optimization methods of godel scheduler: 1. Optimize scheduling algorithms and do data structures refactor; 2. Implement optimistic concurrency under multi-shard architecture to achieve parallel computation; 3. Abstract "batch" scheduling to fully reuse scheduling computation results.

目前，社区中流行的基于Kubernetes的各种任务调度器在性能方面存在一定限制，这限制了它们能处理的集群规模。由于集群规模的限制，通过大规模的共存难以提高资源利用率，而且更多的集群也会带来更大的运维负担。1. 由于调度器及相关组件的瓶颈，最大集群规模无法超过5k个节点；2. 在超过5k个节点的集群中，调度吞吐量无法超过100个Pod/s。 Godel Scheduler是一个基于Kubernetes的分布式高性能调度器，现已开源。在本次演讲中，我们将深入探讨godel调度器的性能优化方法：1. 优化调度算法并进行数据结构重构；2. 在多分片架构下实现乐观并发以实现并行计算；3. 抽象“批量”调度以充分重用调度计算结果。

Speakers

Yuquan Ren

Cloud Native Architect, ByteDance

Yuquan Ren has 10+ years of working experience in the cloud-native field, contributing extensively to open-source projects such as Kubernetes. Currently, he is a tech leader at ByteDance, primarily focusing on the field of orchestration and scheduling.

Bing Li

Senior Software Engineer, Bytedance

Bing Li has participated in the open source community for nearly 3 years. Currently, he is a senior software engineer at ByteDance, focusing on scheduling system performance optimization and system evolution.

Wednesday August 21, 2024 11:00 - 11:35 HKT
Level 1 | Hung Hom Room 2

KubeCon + CloudNativeCon Sessions, Operations + Performance

Experience Level | 内容经验水平 初级 (Beginner)
Language | 语言 中文 (Chinese)

11:00 HKT

Securing the Supply Chain: A Practical Guide to SLSA Compliance from Build to Runtime | 保障供应链安全：从构建到运行的SLSA合规实用指南 - Enguerrand Allamel, Ledger

Wednesday August 21, 2024 11:00 - 11:35 HKT

Level 1 | Hung Hom Room 1

Navigating the complexities of supply chain security might seem intimidating, especially with evolving frameworks like SLSA (Supply-chain Levels for Software Artifacts). This talk introduces beginners to the foundational practices required to secure software from build to runtime using CNCF tools. We'll explore how GitHub Actions can automate build processes, integrate with Cosign for keyless artifact signing, and use Kyverno for runtime policy enforcement. Additionally, we'll discuss how tools like in-toto and Kubescape help manage and verify artifact integrity, providing a holistic view of SLSA compliance in the Kubernetes ecosystem. To enhance security further, we will also briefly discuss the potential integration of Hardware Security Modules (HSMs) into the supply chain. HSMs can offer an added layer of security for key management operations critical to signing processes, ensuring that cryptographic keys are managed securely and are resilient against attack.

在KubeCon的一个会话描述：供应链安全的复杂性可能看起来令人望而却步，尤其是随着像SLSA（软件构件供应链级别）这样不断发展的框架。本次演讲将向初学者介绍使用CNCF工具来确保软件从构建到运行时的基本实践。我们将探讨GitHub Actions如何自动化构建流程，与Cosign集成进行无密钥构件签名，以及使用Kyverno进行运行时策略执行。此外，我们还将讨论像in-toto和Kubescape这样的工具如何帮助管理和验证构件完整性，为Kubernetes生态系统中的SLSA合规性提供全面视角。为了进一步增强安全性，我们还将简要讨论将硬件安全模块（HSMs）集成到供应链中的潜在可能性。HSMs可以为关键管理操作提供额外的安全层，这对签名过程至关重要，确保加密密钥得到安全管理，并且具有抵御攻击的弹性。

Speakers

Enguerrand Allamel

Senior Cloud Security Engineer, Ledger

Enguerrand is a Senior Cloud Security Engineer with experience in Site Reliability Engineering at Ledger since 2022. His work focuses on the security of scalable and reliable cloud systems, leveraging his knowledge of hybrid computing technologies and container orchestration with... Read More →

Wednesday August 21, 2024 11:00 - 11:35 HKT
Level 1 | Hung Hom Room 1

KubeCon + CloudNativeCon Sessions, Security

Experience Level | 内容经验水平 初级 (Beginner)
Language | 语言 英语 (English)

13:50 HKT

Is Your GPU Really Working Efficiently in the Data Center? N Ways to Improve GPU Usage | 您的GPU在数据中心真的高效工作吗？提高GPU使用率的N种方法 - Xiao Zhang, DaoCloud

Wednesday August 21, 2024 13:50 - 14:25 HKT

Level 1 | Hung Hom Room 3

AI has penetrated into various industries, and companies have purchased many expensive AI GPU devices and used them for training and inference. So what is the reality of the use of these devices? Is the usage rate really high? Is the GPU card being monopolized by a large number of applications that are not heavily used? Do these AI devices work efficiently 24/7? This session will combine our mass production practices to summarize N ways to improve the utilization rate of AI equipment, such as * How to avoid monopoly and improve GPU usage through GPU sharing technology * How to improve GPU device usage through co-located in scenes with obvious tides * How to better perform GPU mark group matching training and inference applications to improve GPU usage This session will combine the practical experience of the two open source projects HAMi and Volcano in production, hoping to give everyone a clearer understanding of how to improve GPU usage.

人工智能已经渗透到各个行业，公司购买了许多昂贵的人工智能GPU设备，并将它们用于训练和推理。那么这些设备的使用情况如何呢？使用率真的很高吗？GPU卡是否被大量不常用的应用程序垄断？这些人工智能设备是否能够24/7高效工作？本场演讲将结合我们的大规模生产实践，总结提高人工智能设备利用率的N种方法，例如： * 如何通过GPU共享技术避免垄断并提高GPU使用率 * 如何通过与明显潮汐场景共同使用GPU设备来提高GPU使用率 * 如何更好地执行GPU标记组匹配训练和推理应用程序，以提高GPU使用率本场演讲将结合两个开源项目HAMi和Volcano在生产中的实际经验，希望能让大家更清楚地了解如何提高GPU使用率。

Speakers

xiaozhang

Senior Technical Lead, DaoCloud

- Xiao Zhang is leader of the Container team(focus on infra,AI,Muti-Cluster,Cluster - LCM,OCI) - Kubernetes / Kubernetes-sigs active Contributor、member - Karmada maintainer,kubean maintainer,HAMi maintainer - Cloud-Native Developer - CNCF Open Source Enthusiast. - GithubID: waw... Read More →

Wednesday August 21, 2024 13:50 - 14:25 HKT
Level 1 | Hung Hom Room 3

AI_dev: Open Source GenAI & ML Summit Sessions, Data Engineering + Management

Experience Level | 内容经验水平 初级 (Beginner)
Language | 语言 英语 (English)

14:00 HKT

⚡ Lightning Talk: WASM on Embedded Systems (RTOS) | ⚡ 闪电演讲: 嵌入式系统（RTOS）上的WASM - Han Wu, University of Exeter

Wednesday August 21, 2024 14:00 - 14:05 HKT

Level 1 | Hung Hom Room 1

Web Assembly (WASM) has seen significant success in web applications and is now making inroads into other areas like cloud services and even embedded systems that run Real-time Operating Systems (RTOS), such as Zephyr, RT-Thread, Nuttx, and ESP-IDF. This lighting talk will present different approaches to using WASM on embedded systems. - wasmtime (Arm Linux) - wasm-micro-runtime (RTOS) - wasm3 (Baremetal) The above WASM runtimes offer full support for the WASM core specifications. Additionally, their limited support for the WebAssembly System Interface (WASI) enables access to components such as threads, file systems, and network sockets. Although the WASI specifications that provide access to hardware peripherals such as wasi-i2c, wasi-spi, and wasi-digital-io are still in the early stages of development, the potential advantages in portability, security, and deployment simplicity make WASM a promising choice for embedded systems.

Web Assembly（WASM）在Web应用程序中取得了显著的成功，现在正在进入其他领域，如云服务甚至运行实时操作系统（RTOS）的嵌入式系统，例如Zephyr、RT-Thread、Nuttx和ESP-IDF。这个Lightning Talk将介绍在嵌入式系统中使用WASM的不同方法。 - wasmtime（Arm Linux） - wasm-micro-runtime（RTOS） - wasm3（裸机）上述WASM运行时完全支持WASM核心规范。此外，它们对WebAssembly系统接口（WASI）的有限支持使得可以访问诸如线程、文件系统和网络套接字等组件。尽管提供访问硬件外设的WASI规范，如wasi-i2c、wasi-spi和wasi-digital-io仍处于早期开发阶段，但WASM在可移植性、安全性和部署简易性方面的潜在优势使其成为嵌入式系统的一个有前途的选择。

Speakers

Han

Ph.D. Student, University of Exeter

Ph.D. Student at the University of Exeter in the U.K. for Deep Learning Security in Autonomous Systems. Prior research experience at RT-Thread, LAIX, Xilinx.

Wednesday August 21, 2024 14:00 - 14:05 HKT
Level 1 | Hung Hom Room 1

⚡ Lightning Talks | ⚡ 闪电演讲, Cloud Native Novice

Experience Level | 内容经验水平 初级 (Beginner)
Language | 语言 英语 (English)

14:55 HKT

⚡ Lightning Talk: Tips and Tricks to (Right) Size Your Kubernetes Cluster for Efficiency and Cost Saving | ⚡ 闪电演讲: 为了提高效率和节约成本，调整Kubernetes集群大小的技巧和窍门 - Daniele Polencic, Learnk8s

Wednesday August 21, 2024 14:55 - 15:00 HKT

Level 1 | Hung Hom Room 1

In this session, you will learn how Kubernetes allocates resources in worker nodes and how you can obtain the most out of them by choosing the right kind of limits and requests for your workloads. You will cover some practical tips to allocate the right number of nodes and resources to your cluster: - Should you have larger or smaller nodes? - How reservation affects efficiency and cost savings? - How to "defrag" your cluster to optimize allocations And more.

在这场演讲中，您将学习Kubernetes如何在工作节点中分配资源，以及如何通过为工作负载选择正确的限制和请求来充分利用它们。您将学习一些实用的技巧，来为您的集群分配正确数量的节点和资源： - 您应该选择更大还是更小的节点？ - 预留资源如何影响效率和节约成本？ - 如何“整理”您的集群以优化分配等等。

Speakers

Daniele Polencic

Instructor, Learnk8s

Daniele teaches containers and Kubernetes at Learnk8s. Daniele is a certified Kubernetes administrator by the Linux Foundation. In the last decade, Daniele trained developers for companies in the e-commerce, finance and public sector.

Wednesday August 21, 2024 14:55 - 15:00 HKT
Level 1 | Hung Hom Room 1

⚡ Lightning Talks | ⚡ 闪电演讲, Platform Engineering

Experience Level | 内容经验水平 任意程度 (Any), 初级 (Beginner)
Language | 语言 中文 (Chinese), 英语 (English)

15:35 HKT

Implementing Seamless Connectivity and Service Governance in Multi Kubernetes Cluster with ZTM | 在多个Kubernetes集群中使用ZTM实现无缝连接和服务治理 - Xiaohui Zhang, Flomesh

Wednesday August 21, 2024 15:35 - 16:10 HKT

Level 1 | Hung Hom Room 1

In the evolving cloud-native ecosystem, Kubernetes is vital for microservices. As enterprises adopt multi-cluster Kubernetes setups, securely managing cross-cluster communications becomes challenging due to the limitations of traditional gateways and Ingress solutions. This session explores how ZTM (Zero Trusted Mesh) acts as a bridge across K8s clusters, bypassing traditional gateways and network constraints, thus ensuring zero exposure and boosting security. ZTM uses an HTTP/2-based tunneling mechanism with end-to-end encryption, minimizing public exposure and securing data during transmission. Its design enables quick deployment of cross-cluster communications without altering existing networks or applications, easing management. Furthermore, ZTM integrates with service mesh technologies to provide a secure framework for microservices, supporting service discovery, load balancing, and advanced routing policies, allowing flexible and secure cross-cluster service management.

在不断发展的云原生生态系统中，Kubernetes 对于微服务至关重要。随着企业采用多集群 Kubernetes 设置，由于传统网关和入口解决方案的限制，安全地管理跨集群通信变得具有挑战性。本场演讲探讨了 ZTM（Zero Trusted Mesh）如何作为跨 K8s 集群的桥梁，绕过传统网关和网络限制，从而确保零暴露并提升安全性。 ZTM 使用基于 HTTP/2 的隧道机制进行端到端加密，最大程度减少公开暴露并在传输过程中保护数据安全。其设计能够快速部署跨集群通信，而无需改变现有网络或应用程序，简化管理。此外，ZTM 还与服务网格技术集成，为微服务提供安全框架，支持服务发现、负载均衡和高级路由策略，实现灵活且安全的跨集群服务管理。

Speakers

AddoZhang

Cloud Native Architect, Flomesh

Senior programmer, LFAPAC open source evangelist, CNCF Ambassador, Microsoft MVP, author of the WeChat public account "云原生指北". Years of practical experience in microservices and cloud-native, the main work involves microservices, containers, Kubernetes, DevOps, etc.

Wednesday August 21, 2024 15:35 - 16:10 HKT
Level 1 | Hung Hom Room 1

KubeCon + CloudNativeCon Sessions, Connectivity

Experience Level | 内容经验水平 初级 (Beginner)
Language | 语言 中文 (Chinese)

16:25 HKT

Staying Ahead of Fast-Moving Attackers | 保持领先于快速移动的攻击者 - Aizhamal Nurmamat kyzy, Sysdig

Wednesday August 21, 2024 16:25 - 17:00 HKT

Level 2 | Grand Ballroom 1-2

How to find the right balance between convenience, operational efficiency, and a strong security policy in a world of ephemeral containers? And how can we ensure security at a time when Advanced Persistent Threats (APTs) are more prevalent? In this talk we will present the latest Cloud Native Security & Usage Report findings on critical vulnerabilities inherent in today’s container security practices. We will also demonstrate how a compromised, short-lived container can be an insidious security risk, and what we can do to detect and mitigate those risks in real time using cloud native open source tools.

在一个短暂容器世界中，如何在便利性、运营效率和强大安全政策之间找到合适的平衡？在APT（高级持续性威胁）更加普遍的时代，我们如何确保安全？在这次演讲中，我们将介绍最新的云原生安全和使用报告发现，揭示当今容器安全实践中存在的关键漏洞。我们还将演示一个被 compromise 的短暂容器如何成为一个隐蔽的安全风险，以及我们如何使用云原生开源工具实时检测和减轻这些风险。

Speakers

Aizhamal Nurmamat kyzy

Director, DevRel, Sysdig

Aizhamal is a Director of DevRel at Sysdig where she focuses on education around security and open source. Previously she worked at Google's OSPO where she helped build open source communities in cloud native and data analytics ecosystems.

Wednesday August 21, 2024 16:25 - 17:00 HKT
Level 2 | Grand Ballroom 1-2

KubeCon + CloudNativeCon Sessions, Security

Experience Level | 内容经验水平 初级 (Beginner)
Language | 语言 英语 (English)

17:15 HKT

How to Manage Database Clusters Without a Dedicated Operator | 如何在没有专门Operator的情况下管理数据库集群 - Shanshan Ying, ApeCloud & Shun Ding, China Mobile Cloud

Wednesday August 21, 2024 17:15 - 17:50 HKT

Level 1 | Hung Hom Room 2

As Kubernetes becomes integral to cloud-native environments, more organizations are deploying database services on K8S, facing significant challenges. Integrating new database engines typically requires developing a dedicated Kubernetes operator that manages not only resource provisioning but also essential maintenance tasks like high availability, backup & restore, and configuration management. This session introduces a universal operator framework that supports various database engines, enabling rapid, minimal-code integration. We will present a case study from China Mobile Cloud on integrating a new cloud-native database engine into K8S using this framework, achieved with minimal coding and reduced time investment, bypassing the extensive Golang coding usually required for developing a dedicated operator.

随着Kubernetes成为云原生环境中不可或缺的一部分，越来越多的组织在K8S上部署数据库服务，面临着重大挑战。集成新的数据库引擎通常需要开发一个专门的Kubernetes operator，管理资源提供以及高可用性、备份和恢复、配置管理等重要维护任务。本场演讲将介绍一个支持各种数据库引擎的通用operator框架，实现快速、最小代码集成。我们将从中国移动云的一个案例研究中介绍如何使用这个框架将新的云原生数据库引擎集成到K8S中，通过最小的编码和减少时间投入来实现，避免通常需要开发专门operator所需的大量Golang编码。

Speakers

Shanshan Ying

Maintainer, ApeCloud

Shanshan is currently a maintainer of KubeBlocks by ApeCloud. Before joining ApeCloud, she worked in Aliyun Database Group for years. She received her PhD degree from National University of Singapore.

Shun Ding

Senior Systems Architect, China Mobile Cloud

Shun is a Senior Systems Architect at China Mobile Cloud, leading the design, development, and deployment of next-generation Kubernetes-based large-scale database managing service. With over a decade of experience in cloud computing and database technologies, Shun has extensive expertise... Read More →

Wednesday August 21, 2024 17:15 - 17:50 HKT
Level 1 | Hung Hom Room 2

KubeCon + CloudNativeCon Sessions, Operations + Performance

Experience Level | 内容经验水平 初级 (Beginner)
Language | 语言 英语 (English)

10:05 HKT

Keynote: Supporting Large-Scale and Reliability Testing in Kubernetes using KWOK | 主论坛演讲: 支持在Kubernetes中使用KWOK进行大规模和可靠性测试 - Yuan Chen, NVIDIA & Shiming Zhang, DaoCloud

Thursday August 22, 2024 10:05 - 10:20 HKT

Level 2 | Grand Ballroom 1-2

Kubernetes is the de facto platform for running workloads at scale. This talk will present KWOK (https://kwok.sigs.k8s.io/), an open-source toolkit that enables the creation and testing of large-scale Kubernetes clusters with minimal resources, even on a laptop.
Shiming Zhang, the creator and maintainer of KWOK, and Yuan Chen, an engineer at NVIDIA GPU Cloud, will outline KWOK's capabilities to generate and manage a large number of virtual nodes that simulate Kubelet APIs and mimic real nodes, allowing for workload deployment and testing. They will discuss practical use cases of KWOK.

The talk will then introduce KWOK's recent enhancements for reliability and fault-tolerance testing, showcasing its ability to simulate failures by injecting targeted faults into nodes and pods. Through examples and demos, the talk will demonstrate how KWOK can be used for reliability testing and evaluating fault-tolerance mechanisms, ultimately improving workload resilience in Kubernetes.

Kubernetes是运行大规模工作负载的事实标准平台。本次演讲将介绍KWOK（https://kwok.sigs.k8s.io/），这是一个开源工具包，可以利用极少的资源（甚至在笔记本电脑上）创建和测试大规模Kubernetes集群。

KWOK的创始人和维护者张世明，以及NVIDIA GPU Cloud的工程师陈源，将详细阐述KWOK的功能，包括生成和管理大量模拟Kubelet API和真实节点的虚拟节点，从而支持工作负载的部署和测试。他们将讨论KWOK的实际使用案例。

演讲还将介绍KWOK最近针对可靠性和容错性测试的增强功能，展示其通过向节点和Pod注入有针对性的故障来模拟故障的能力。通过示例和演示，演讲将展示如何利用KWOK进行可靠性测试和评估容错机制，从而最终提升Kubernetes中工作负载的弹性能力。

Speakers

Yuan Chen

Principal Software Engineer, NVIDIA

Yuan Chen is a Principal Software Engineer at NVIDIA, working on building NVIDIA GPU Cloud. He served as a Staff Software Engineer at Apple from 2019 to 2024, where he contributed to the development of Apple's Kubernetes infrastructure. Yuan has been an active code contributor to... Read More →

Shiming Zhang

Software Engineer, DaoCloud

Shiming Zhang is a contributor to Kubernetes with the main focus on scalability, performance, reliability, and testing, he gained experience and contributed to many Kubernetes features and most of its components.

Thursday August 22, 2024 10:05 - 10:20 HKT
Level 2 | Grand Ballroom 1-2

Keynote Sessions | 主论坛演讲, Operations + Performance

Experience Level | 内容经验水平 初级 (Beginner)
Language | 语言 英语 (English)

11:00 HKT

Unlocking the Power of Kubernetes: AI-Driven Innovations for Next-Gen Infrastructure | 释放 Kubernetes 的力量：面向下一代基础设施的 AI 驱动创新 - Brandon Kang, Akamai Technologies

Thursday August 22, 2024 11:00 - 11:35 HKT

Level 1 | Hung Hom Room 3

My session is about dynamic synergy between Kubernetes and AI, unveiling a transformative paradigm shift in modern infrastructure management. The presentation unveils how Kubernetes serves as an enabler for deploying and scaling AI workloads efficiently, optimizing resource utilization, and ensuring unparalleled scalability. Delving deeper, it explores the realm of AI-powered automation, showcasing how intelligent algorithms enhance auto-scaling, workload optimization, and predictive maintenance within Kubernetes clusters. Moreover, it sheds light on the crucial aspect of security, elucidating how AI-driven measures bolster threat detection and anomaly identification, fortifying Kubernetes environments against potential risks. This presentation beckons organizations to embrace the convergence of Kubernetes and AI, unlocking boundless possibilities to redefine infrastructure management and propel towards unprecedented efficiency and resilience.

我的演讲是关于 Kubernetes 和人工智能之间的动态协同作用，揭示了现代基础设施管理中的转变范式。演示展示了 Kubernetes 如何作为部署和扩展人工智能工作负载的促进者，有效优化资源利用率，并确保无与伦比的可扩展性。更深入地探讨了基于人工智能的自动化领域，展示了智能算法如何增强 Kubernetes 集群内的自动扩展、工作负载优化和预测性维护。此外，它还阐明了安全的关键方面，阐明了人工智能驱动的措施如何加强威胁检测和异常识别，加固 Kubernetes 环境抵御潜在风险。这个演示呼吁组织 embrace Kubernetes 和人工智能的融合，解锁无限可能性，重新定义基础设施管理，并朝着前所未有的效率和韧性迈进。

Speakers

Brandon Kang

Principal Technical Solutions Architect, Akamai Technologies

Brandon Kang is a Cloud Specialist at Akamai Technologies, where he oversees cloud computing projects across the APJ markets, including China. Before his tenure at Akamai, Brandon was a software engineer at Samsung, a program manager at Microsoft, and a service platform expert at... Read More →

Thursday August 22, 2024 11:00 - 11:35 HKT
Level 1 | Hung Hom Room 3

AI_dev: Open Source GenAI & ML Summit Sessions, Edge + Distributed AI

Experience Level | 内容经验水平 初级 (Beginner)
Language | 语言 英语 (English)

11:50 HKT

Redefining Service Mesh: Leveraging EBPF to Optimize Istio Ambient Architecture and Performance | 重新定义服务网格：利用eBPF优化Istio环境架构和性能 - Yuxing Zeng, Alibaba Cloud

Thursday August 22, 2024 11:50 - 12:25 HKT

Level 1 | Hung Hom Room 1

Istio Ambient separates the L4/L7 functions found in the traditional sidecar model and introduces the ztunnel component, which implement the L4 network load balancing and secure zero-trust. However, as ztunnel is deployed at the node level with DaemonSet, any malfunction or anomaly in ztunnel may impact the traffic of all mesh-related pods under that node. Furthermore, performance tests of Ambient Mesh have not delivered the anticipated outcomes; ztunnel often becomes a performance bottleneck. These factors make it challenging to apply Ambient Mesh in production environments. it appears that we require a more optimized and practical implementation solution. This session will share: 1. An introduction to the architecture of Istio Ambient Mesh, along with current known issues with the existing implement. 2. using eBPF to implement zero-trust and L4 network traffic capabilities, enhancing the stability of the Mesh network, and significantly improving overall performance.

Istio Ambient将传统的边车模型中发现的L4/L7功能分离，并引入了ztunnel组件，实现了L4网络负载均衡和安全的零信任。然而，由于ztunnel部署在节点级别的DaemonSet上，ztunnel中的任何故障或异常可能会影响该节点下所有与网格相关的Pod的流量。此外，Ambient Mesh的性能测试并未达到预期的结果；ztunnel经常成为性能瓶颈。这些因素使得在生产环境中应用Ambient Mesh变得具有挑战性。看起来我们需要一个更优化和实用的实现解决方案。本次会话将分享： 1. Istio Ambient Mesh架构的介绍，以及现有实现中已知的问题。 2. 使用eBPF实现零信任和L4网络流量功能，增强Mesh网络的稳定性，并显著提高整体性能。

Speakers

Jesse Zeng

Technical Expert, Alibaba Cloud

Yuxing Zeng is a technical expert in the Container Service Team at Alibaba Cloud. He is also a Istio Member、Envoy Contributor. He has rich experience in cloud native fields such as Kubernetes、Istio、 Envoy, etc.

Thursday August 22, 2024 11:50 - 12:25 HKT
Level 1 | Hung Hom Room 1

KubeCon + CloudNativeCon Sessions, Connectivity

Experience Level | 内容经验水平 初级 (Beginner)
Language | 语言 英语 (English)

13:50 HKT

Unified Management, Continuity, Compliance in Multi-Clouds with Service Mesh | 在多云环境中通过服务网格实现统一管理、连续性和合规性 - Kebe Liu, DaoCloud

Thursday August 22, 2024 13:50 - 14:25 HKT

Level 1 | Hung Hom Room 1

In multi-cloud and hybrid cloud architectures, enterprises face challenges like inter-cloud communication, traffic management, application orchestration, data security, and compliance. Service mesh technology offers a unified approach for managing service interactions, enhancing security, and ensuring data compliance. Istio, a leading service mesh project, is particularly effective in multi-cloud and hybrid cloud environments. It provides seamless network connectivity across various architectures, ensuring reliable and secure communication. Additionally, integrating Istio with Karmada enables efficient application scheduling across these complex environments. Karmada allows for smooth orchestration of workloads across different cloud platforms, enhancing the flexibility and scalability of cloud-native applications. I aim to share practical insights and experiences, especially from China, to inspire and provide strategic perspectives in navigating these technological landscapes.

在多云和混合云架构中，企业面临诸如云间通信、流量管理、应用编排、数据安全和合规性等挑战。服务网格技术提供了统一的管理服务交互方式，增强安全性，并确保数据合规性。作为领先的服务网格项目，Istio在多云和混合云环境中特别有效。它提供了跨不同架构的无缝网络连接，确保可靠和安全的通信。此外，将Istio与Karmada集成，可以实现在这些复杂环境中高效的应用调度。Karmada允许在不同云平台上平稳地编排工作负载，增强云原生应用的灵活性和可扩展性。我旨在分享实用的见解和经验，特别是来自中国，以激发并提供在这些技术领域中导航的战略视角。

Speakers

Kebe Liu

Senior software engineer, DaoCloud

Member of Istio Steering Committee, focused on cloud-native and Istio, eBPF and other areas in recent years. Founder of Merbridge project.

Thursday August 22, 2024 13:50 - 14:25 HKT
Level 1 | Hung Hom Room 1

KubeCon + CloudNativeCon Sessions, Connectivity

Experience Level | 内容经验水平 初级 (Beginner)
Language | 语言 中文 (Chinese)

14:40 HKT

Find Your Own Personal Tutor for the Study of Kubernetes | 为学习Kubernetes找到适合您的个人导师 - Hoon Jo, Megazone

Thursday August 22, 2024 14:40 - 15:15 HKT

Level 1 | Hung Hom Room 2

Kubernetes novice users ask questions to stackoverflow or community or friends :) when they encounter the problem. However it needs to explain my environment and the background information. Even though it is not a guaranteed answer from someone. Thus I suggest to use K8sGPT with ollama to leverage the lack of knowledge at this moment. Furthermore, k8sGPT provides interactive mode that is able to ask continuing questions until I receive enough answers. Plus it could be helpful to ask other language who is not familiar with English. (Mostly it is big concern from the beginning of the stage) I highly recommend using K8sGPT to study who is a newcomer for soft landing in Kubernetes world.

在KubeCon上，我们将讨论Kubernetes新手用户在遇到问题时通常会向stackoverflow、社区或朋友提问的情况。然而，我们需要解释我的环境和背景信息。虽然并不能保证会得到答案，但我建议使用K8sGPT与ollama来弥补当前知识的不足。此外，k8sGPT提供交互模式，可以持续提问直到我得到足够的答案。此外，对于不熟悉英语的人来说，询问其他语言可能会有所帮助（这在刚开始阶段时是一个大问题）。我强烈推荐使用K8sGPT来帮助新手顺利进入Kubernetes世界。

Speakers

Hoon Jo

Cloud Solutions Architect | Cloud Native Engineer,, Megazone

Hoon Jo is Cloud Solutions Architect as well as Cloud Native engineer at Megazone. He has many times of speaker experience for cloud native technologies. And spread out Cloud Native Ubiquitous in the world. He wrote 『Python for System/Network Administrators』 (Wikibooks, 2017... Read More →

Thursday August 22, 2024 14:40 - 15:15 HKT
Level 1 | Hung Hom Room 2

KubeCon + CloudNativeCon Sessions, Cloud Native Novice

Experience Level | 内容经验水平 初级 (Beginner)
Language | 语言 英语 (English)

15:35 HKT

Revolutionizing Scientific Simulations with Argo Workflows | 用Argo工作流彻底改变科学模拟 - ShaungKun Tian, Alibaba Cloud & 建翔孙, 北京深势科技有限公司

Thursday August 22, 2024 15:35 - 16:10 HKT

Level 1 | Hung Hom Room 1

DP Technology provides scientific simulation platforms for research in biomedicine, energy, materials and other industries. Science simulation workflows are inherently complex and resource-intensive, and manual deployment is often prone to errors. After adopting Argo workflows to orchestrate science simulation, we get productivity 100% improvement. In this talk, we will introduce why chose Argo Workflow, how to orchestrate large-scale tasks of science simulation, how to make whole system scalability and reliability. Specially, we will share best practice about how manage super large workflow (thousands of tasks), how to do reasonable workflow retry, how to use memorization to reduce runtime and compute cost, how to interact with HPC systems. We also made contributions to Argo community to enhance functionalities and improve reliability. Additionally, we'll introduce DFlow, our open-source Python SDK designed for the seamless orchestration of scientific simulations with Argo Workflows.

DP Technology为生物医药、能源、材料等行业的研究提供科学模拟平台。科学模拟工作流程本质上复杂且资源密集，手动部署往往容易出错。采用Argo工作流程来编排科学模拟后，我们的生产力提高了100%。在本次演讲中，我们将介绍为什么选择Argo工作流程，如何编排大规模科学模拟任务，如何实现整个系统的可扩展性和可靠性。特别是，我们将分享如何管理超大型工作流程（数千个任务），如何合理重试工作流程，如何使用记忆化来减少运行时间和计算成本，如何与HPC系统交互。我们还为Argo社区做出了贡献，以增强功能性和提高可靠性。此外，我们还将介绍DFlow，我们的开源Python SDK，旨在与Argo工作流程无缝协同编排科学模拟。

Speakers

建翔孙

软件工程师, 北京深势科技有限公司

I once built a machine learning platform at Kuaishou, and currently, I am involved in scheduling scientific computing tasks at DP Technology, as well as constructing workflow platforms. I specialize in the field of cloud-native development.

Thursday August 22, 2024 15:35 - 16:10 HKT
Level 1 | Hung Hom Room 1

KubeCon + CloudNativeCon Sessions, Platform Engineering

Experience Level | 内容经验水平 初级 (Beginner)
Language | 语言 英语 (English)

15:35 HKT

Phippy’s Field Guide to Wasm | Phippy的Wasm指南 - Karen Chu, Fermyon & Matt Butcher, Fermyon Technologies

Thursday August 22, 2024 15:35 - 16:10 HKT

Level 1 | Hung Hom Room 5

The creators of the original Illustrated Children’s Guide to Kubernetes have written a fourth book, this time focused on the emerging technology that is WebAssembly, one of the fastest growing cloud native trends. As with previous books, we broach a complex technical topic with a fun and friendly format designed for all skill levels. On their camping trip with Blossom the Wasm Possum, Phippy and Zee’s adventures illustrate the basics of Wasm, introduce key terminology, and frame how it compliments existing cloud technologies like containers and Kubernetes. In the first half of the talk, we will do a reading of the book in Mandarin. We will then follow up (in English) with a technical overview of Wasm, latest updates to the ecosystem, and details on where to find the community.

原《插图儿童 Kubernetes 指南》的创作者们已经写了第四本书，这次的焦点是新兴技术 WebAssembly，这是增长最快的云原生趋势之一。与之前的书籍一样，我们以有趣友好的格式涉及复杂的技术主题，适合各种技能水平的读者。在他们与 Wasm 负鼠 Blossom 一起露营的旅行中，Phippy 和 Zee 的冒险展示了 Wasm 的基础知识，介绍了关键术语，并阐述了它如何与容器和 Kubernetes 等现有云技术相辅相成。在讲座的前半部分，我们将用普通话朗读这本书。然后我们将用英语进行技术概述，介绍 Wasm 生态系统的最新更新，并详细介绍社区的位置。

Speakers

Karen Chu

Head of Community, Fermyon

Karen Chu is the Head of Community at Fermyon Technologies. Having participated in the cloud native community since 2015, she is a CNCF Ambassador, Helm community manager/maintainer, emeritus Kubernetes Code of Conduct Committee member, meet-up organizer, and conference organizer... Read More →

Matt Butcher

CEO, Fermyon Technologies

Matt Butcher (CEO) is a founder of Fermyon. He is one of the original creators of Helm, Brigade, CNAB, OAM, Glide, and Krustlet. He has written or co-written many books, including "Learning Helm" and "Go in Practice." He is a co-creator of the "Illustrated Children’s Guide to Kubernetes... Read More →

Thursday August 22, 2024 15:35 - 16:10 HKT
Level 1 | Hung Hom Room 5

Open Source Summit Sessions, Programming Languages and Toolchains

Experience Level | 内容经验水平 初级 (Beginner)
Language | 语言 英语 (English)

10:35 HKT

Breaking Boundaries: TACC as an Unified Cloud-Native Infra for AI + HPC | 打破界限：TACC作为AI + HPC统一云原生基础设施 - Peter Pan, DaoCloud & Kaiqiang Xu, Hong Kong University of Science and Technology

Friday August 23, 2024 10:35 - 11:10 HKT

Level 1 | Hung Hom Room 3

Large AI models are driving significant investment in GPU clusters. Yet, managing these clusters is hard: Slurm-based HPC setups lack of management granularity and stability, while Kubernetes poses usability challenges for AI users. This talk introduces TACC, an AI infra management solution that bridges the advantages of both K8S and Slurm setups. This is a joint-work from computer system researchers at HKUST and leading CNCF contributors at DaoCloud. TACC manages a large-scale cluster at HKUST that supports over 500 active researchers since 2020. In this talk, we share our five-year journey with TACC, covering: * [User Experience] A seamless UI for job submissions and management, supporting both container and Slurm format, all on the same backbone * [Resource Management] Multi-tenant allocation with configurable strategies, using CNCF HAMi and Kueue * [Performance and Scalability] A robust distributed infrastructure with networked storage and RDMA, via CNCF SpiderPool,Fluid...

大型AI模型正在推动GPU集群的重大投资。然而，管理这些集群很困难：基于Slurm的HPC设置缺乏管理粒度和稳定性，而Kubernetes对AI用户存在可用性挑战。本次演讲介绍了TACC，这是一种AI基础设施管理解决方案，可以结合K8S和Slurm设置的优势。这是香港科技大学的计算机系统研究人员与DaoCloud领先的CNCF贡献者共同合作的成果。 TACC自2020年以来管理着香港科技大学支持超过500名活跃研究人员的大规模集群。在本次演讲中，我们分享了与TACC一起的五年历程，涵盖以下内容： * [用户体验] 无缝的UI界面用于作业提交和管理，支持容器和Slurm格式，均在同一基础上 * [资源管理] 多租户分配与可配置策略，使用CNCF HAMi和Kueue * [性能和可扩展性] 强大的分布式基础设施，具有网络存储和RDMA，通过CNCF SpiderPool，Fluid...

Speakers

Peter Pan

VP of R&D Engineering, DaoCloud

├ DaoCloud R&D Engineering VP├ CNCF wg-AI (AI Working-Group) member├ Maintainer of a few CNCF projects (GithubID: panpan0000): CloudTTY, KuBean, HwameiStor├ Public Tech Events:└─ 2023 KubeCon SH Speaker (https://sched.co/1PTFI)└─ 2023 KubeCon EU Program Committee... Read More →

Kaiqiang Xu

Researcher, Hong Kong University of Science and Technology

Hong Kong University of Science and Technology

Friday August 23, 2024 10:35 - 11:10 HKT
Level 1 | Hung Hom Room 3

AI_dev: Open Source GenAI & ML Summit Sessions, Foundations + Frameworks + Tools for Machine Learning

Experience Level | 内容经验水平 初级 (Beginner)
Language | 语言 英语 (English)

10:35 HKT

A Year in the Life of a Developer in the Era of Developer Portals: Navigating Backstage | 开发者在开发者门户时代的一年生活：导航Backstage - Helen Greul, Spotify

Friday August 23, 2024 10:35 - 11:10 HKT

Level 1 | Hung Hom Room 1

In today's rapidly evolving landscape of software development, the role of developer portals has become indispensable. This presentation delves into the experiences of developers over the course of a year, exploring the transformative impact of Backstage developer portal on their workflows, collaboration, and overall productivity based on case studies from existing adopters of Backstage. Through a comprehensive exploration of real-world scenarios, this talk offers insights into the daily challenges faced by developers and how Backstage empowers them to overcome these hurdles. From streamlined onboarding processes to simplified access to internal services and documentation, attendees will gain a deeper understanding of the multifaceted benefits that Backstage brings to developer teams. Moreover, we'll discuss best practices for leveraging Backstage to foster a culture of innovation, collaboration, and continuous improvement.

在当今快速发展的软件开发领域，开发者门户的作用变得不可或缺。本次演讲将深入探讨开发者在一年时间内的经验，通过现有Backstage采用者的案例研究，探讨Backstage开发者门户对他们的工作流程、协作和整体生产力的转变影响。通过对现实场景的全面探讨，本次演讲将为参与者提供洞察开发者面临的日常挑战，以及Backstage如何赋予他们克服这些障碍的能力。从简化入职流程到简化访问内部服务和文档，参与者将更深入地了解Backstage为开发团队带来的多方面好处。此外，我们还将讨论利用Backstage促进创新、协作和持续改进文化的最佳实践。

Speakers

Helen Greul

Head of Engineering for Backstage, Spotify

Helen is an engineering leader, speaker and a strong advocate for creating developer ecosystems that empower teams to thrive. Her journey has taken her from hands-on coding to steering engineering and platform teams, providing her with a holistic perspective on the challenges and... Read More →

Friday August 23, 2024 10:35 - 11:10 HKT
Level 1 | Hung Hom Room 1

KubeCon + CloudNativeCon Sessions, Platform Engineering

Experience Level | 内容经验水平 初级 (Beginner)
Language | 语言 英语 (English)

11:25 HKT

Beyond Statefulset: Containerize Your Enterprise Stateful Applications in Practice | 超越StatefulSet：实践中将企业有状态应用容器化 - Mingshan Zhao, Alibaba Cloud & Vec Sun, xiaohongshu

Friday August 23, 2024 11:25 - 12:00 HKT

Level 1 | Hung Hom Room 1

Kubernetes provides StatefulSet to manage stateful services, but it is far from enough to run enterprise stateful applications in practice. For example: how does Zookeeper accomplish leader election, and how does MQ implement configuration hot loading? How to do daily operation and maintenance of the database? Many practitioners resort to operators that manages pod directly e.g. KubeBlocks, for specific applications e.g. database, yet they are not general enough for other stateful applications. OpenKruise provides several stateful features that are missing in native StatefulSet, such as in-place resource and volume resizing, progressive Configmap & Secret hot update and container operation channel. Teams from Alibaba and Xiaohongshu will share their lessons to build operators and platforms for general stateful apps and containerize database and middleware with a scale of hundreds of thousands of pods.

Kubernetes提供了StatefulSet来管理有状态服务，但实际上要运行企业级有状态应用还远远不够。例如：Zookeeper如何完成领导者选举，MQ如何实现配置热加载？如何进行数据库的日常运维？许多从业者借助直接管理pod的运营商，例如KubeBlocks，针对特定应用程序，例如数据库，但它们并不足够通用以适用于其他有状态应用程序。 OpenKruise提供了一些在原生StatefulSet中缺失的有状态功能，例如原地资源和卷大小调整，渐进式Configmap和Secret热更新以及容器操作通道。来自阿里巴巴和小红书的团队将分享他们构建运营商和平台以适用于通用有状态应用程序，并将数据库和中间件容器化的经验，规模达数十万个pod。

Speakers

Mingshan Zhao

Senior R&D Engineer, Alibaba Cloud

Senior R&D Engineer of AliCloud, Maintainer of OpenKruise community, has long been engaged in the research and development of cloud native, containers, scheduling and other fields; core R&D member of Alibaba's one million container scheduling system, and many years of experience in... Read More →

Vec Sun

software engineer, xiaohongshu

Sunweixiang has previously worked in the Alibaba Cloud container team as software engineer and is a contributor to the OpenKruise community's main, Karmada, and other communities. He is deeply involved in container application orchestration, multi-cluster.

Friday August 23, 2024 11:25 - 12:00 HKT
Level 1 | Hung Hom Room 1

KubeCon + CloudNativeCon Sessions, Platform Engineering

Experience Level | 内容经验水平 初级 (Beginner)
Language | 语言 中文 (Chinese)

11:25 HKT

Rollout Patterns: Smoothly Migrating and Rolling Out Your Microservices | 部署模式：平稳迁移和部署您的微服务 - Tim Xiao, DaoCloud & Wu Chenhui, AS.Watson TechLab

Friday August 23, 2024 11:25 - 12:00 HKT

Level 1 | Hung Hom Room 7

At Watsons, most of their services are built on Dubbo. Now, they aim to utilize delivery tools like Argo CD and Argo Rollouts to automatically and securely deliver their services. However, they have encountered complexities beyond what Argo Rollouts assumes. We will summarize these patterns and demonstrate how to handle them, including: - Pattern 1: One service at a time. - Pattern 2: Multiple services, each forward-compatible. - Pattern 3: Multiple services with version dependency.

在Watsons，他们的大多数服务都是基于Dubbo构建的。现在，他们希望利用Argo CD和Argo Rollouts等交付工具来自动和安全地交付他们的服务。然而，他们遇到了超出Argo Rollouts假设的复杂性。我们将总结这些模式，并演示如何处理它们，包括： - 模式1：一次一个服务。 - 模式2：多个服务，每个都是向前兼容的。 - 模式3：具有版本依赖性的多个服务。

Speakers

旸肖

Developer, DaoCloud

Served as DevOps platform Principle Engineer in DaoCloud, participated in community projects including argo-cd, argo-rollouts, kubevela and other community projects, and has more than 5 years of kubernetes platform development experience.

Wu Chenhui

architecture, AS.Watson TechLab

I have nearly 30 years of experience in software development and architecture design, and 5 years of experience in k8s, responsible for k8s related architecture design of Watsons Group

Friday August 23, 2024 11:25 - 12:00 HKT
Level 1 | Hung Hom Room 7

KubeCon + CloudNativeCon Sessions, SDLC (Software Development Lifecycle)

Experience Level | 内容经验水平 初级 (Beginner)
Language | 语言 中文 (Chinese)

11:25 HKT

How Does KubeEdge Build the Tunnel Which Is Secure, Trusted, and Adaptable to Edge Networks | KubeEdge如何构建适应边缘网络的安全可信隧道 - Wei Hu, DaoCloud

Friday August 23, 2024 11:25 - 12:00 HKT

Level 1 | Hung Hom Room 5

Edge Computing makes the connection broader, faster and more agile, meanwhile it also brings the threat of cyberattacks to the edge of the network, which also puts forward higher requirements for the safety at the edge side. In addition, due to any forms like Internet, 5G, WIFI and other forms are possible, the network environment will be complex and the quality can't be guaranteed in the edge scnee. Therefore, supporting weak network environments which is also a challenge at edge site. KubeEdge is a cloud-edge collaborative architecture project for Kubernetes native edge computing. KubeEdge uses its own trusted tunnel to ensure the security of data transmission, it verifies, encrypts and authenticates all communications in this tunnel. This tunnel ensures data accessibility through QoS and provides a QUIC protocol to improve the performance of network reordering in weak networks. We will share how the tunnel of KubeEdge achieves these goals in this session.

边缘计算使连接更广泛、更快速、更灵活，同时也将网络威胁带到了边缘，这也对边缘安全提出了更高的要求。此外，由于互联网、5G、WIFI等各种形式可能存在，边缘场景中的网络环境将变得复杂，质量无法保证。因此，支持弱网络环境也是边缘场景中的一个挑战。 KubeEdge是一个针对Kubernetes原生边缘计算的云边协作架构项目。KubeEdge使用自己的可信隧道来确保数据传输的安全性，它验证、加密和认证该隧道中的所有通信。该隧道通过QoS确保数据可访问性，并提供QUIC协议来改善弱网络中的网络重排序性能。在本场演讲中，我们将分享KubeEdge隧道如何实现这些目标。

Speakers

炜胡

Senior Software Engineer, DaoCloud

Wei Hu is a Senior Software Engineer at DaoCloud, currently working on Edge Computing Team. He is a maintainer of KubeEdge project and a regular contributor to it. He has rich experience in cloud-edge collaboration. He has given several speeches on the topic of edge computing at other... Read More →

Friday August 23, 2024 11:25 - 12:00 HKT
Level 1 | Hung Hom Room 5

Open Source Summit Sessions, Networking + Edge Computing

Experience Level | 内容经验水平 初级 (Beginner)
Language | 语言 中文 (Chinese)

13:20 HKT

Constructing the 10x Efficiency of Cloud-Native AI Infrastructure | 如何让你的 AI 底座效能提升 10 倍？ - Peter Pan, DaoCloud & 秋萍戴, daocloud

Friday August 23, 2024 13:20 - 13:55 HKT

Level 1 | Hung Hom Room 2

Enterprises keep invested in AI. But once GPU are installed in a data center, a challenge arises: how to construct an "AI cloud" atop bare-metal. Even when K8S is recognized as the foundational infrastructure for AI, But K8S only is merely the initial step. Organizations may face challenges: - Maximizing GPU utilization - Unifying multi-arch accelerators/GPUs (k8s DRA) - Organization quotas and cost management - Resource isolation among organizations - Smarter scheduling, tiered GPU allocation, task prioritization.. - Sharing GPU clusters between VMs & containers - Harnessing the full potential of high-speed networks , Storage optimization and dataset orchestration Leveraging open source stacks in Linux Foundation and CNCF, we've experience in building AI clouds for IDC or internal usage. We can share experiences to empower communities' journey towards constructing the 10x efficiency of cloud-native AI. Refer to `Additional resources` chapter for more details

企业继续投资于人工智能。但是一旦在数据中心安装了GPU，就会面临一个挑战：如何在裸金属之上构建一个“AI云”。即使K8S被认为是AI的基础基础设施，但K8S只是一个起步。组织可能面临的挑战包括： - 最大化GPU利用率 - 统一多架构加速器/GPU（k8s DRA） - 组织配额和成本管理 - 组织之间的资源隔离 - 更智能的调度，分层GPU分配，任务优先级... - 在虚拟机和容器之间共享GPU集群 - 充分利用高速网络的潜力，优化存储和数据集编排利用Linux基金会和CNCF中的开源堆栈，我们在为IDC或内部使用构建AI云方面有经验。我们可以分享经验，以赋予社区构建云原生AI的效率提升10倍的旅程。有关更多详细信息，请参考“附加资源”章节。

Speakers

Peter Pan

VP of R&D Engineering, DaoCloud

秋萍戴

product mananger, daocloud

QiuPing Dai is a senior Technology Product Manager at DaoCloud for 5 years and involved in Cloud Computing ( including Kubernetes Computing, Storage, Network) development work. Before that, Qiuping worked at IBM for Cloud Computing. QiuPing is interested in Storage, Network , Scheduling... Read More →

Friday August 23, 2024 13:20 - 13:55 HKT
Level 1 | Hung Hom Room 2

AI_dev: Open Source GenAI & ML Summit Sessions, Foundations + Frameworks + Tools for Machine Learning

Experience Level | 内容经验水平 初级 (Beginner)
Language | 语言 英语 (English)

14:10 HKT

Unveiling the Future: Nurturing Openness in AI Development | 揭示未来：培育人工智能开放性发展 - Anni Lai, Futurewei & Mer Joyce, Do Big Good LLC

Friday August 23, 2024 14:10 - 14:45 HKT

Level 1 | Hung Hom Room 3

In the rapidly evolving landscape of AI, the concept of openness emerges as a cornerstone for ethical, accountable, and sustainable development. This talk delves into the significance of fostering openness in AI endeavors, exploring two groundbreaking efforts: the Open Source AI Definition led by the Open Source Initiative (OSI) and Model Openness Framework (MOF) introduced by LF AI & Data Generative AI Commons. Through the lens of the OSI's definition co-design process, we'll navigate the evolving landscape of Open Source AI, deciphering its potential to democratize access to cutting-edge technology while fortifying principles of inclusivity and collaboration. We'll unravel the transformative potential of the MOF to foster transparency and trust in AI models. By elucidating the core tenets of the framework and the definition, we'll illuminate pathways for advancing responsible AI development.

在人工智能快速发展的领域中，开放性的概念成为道德、负责任和可持续发展的基石。本次演讲深入探讨了在人工智能努力中培育开放性的重要性，探索了两项开创性的工作：由开源倡议组织（OSI）领导的开源人工智能定义和LF AI & Data生成人工智能共同体引入的模型开放性框架（MOF）。通过OSI定义的共同设计过程，我们将探索开源人工智能不断发展的领域，解读其潜力，使人们能够民主化获得尖端技术，同时巩固包容性和合作原则。我们将揭示MOF在促进人工智能模型透明度和信任方面的转变潜力。通过阐明框架和定义的核心原则，我们将阐明推进负责任人工智能发展的途径。

Speakers

Anni Lai

Head of Open Source Operations, Chair of Generative AI Commons, LF AI & Data, Futurewei

Anni drives Futurewei’s open source (O.S.) governance, process, compliance, training, project alignment, and ecosystem building. Anni has a long history of serving on various O.S. boards such as OpenStack Foundation, LF CNCF, LF OCI, LF Edge, and is on the LF OMF board and LF Europe... Read More →

Mer Joyce

Founder, Do Big Good LLC

Mer Joyce (she/her) is the founder of the co-design firm Do Big Good and is the facilitator of the Open Source Initiative's consultative process to co-design the Open Source AI Definition (OSAID). She has over a decade of international experience at the intersection of research, tech... Read More →

Friday August 23, 2024 14:10 - 14:45 HKT
Level 1 | Hung Hom Room 3

AI_dev: Open Source GenAI & ML Summit Sessions, Responsible AI: Ethics + Security + Governance in AI

Experience Level | 内容经验水平 初级 (Beginner)
Language | 语言 英语 (English)