with the use of groundbreaking technologies, to enable decision-making at the point of data collection. Fast, insight-driven decision-making in highly dynamic and dangerous conditions is, more often analysis at the point of collection, supplying imme- diate insight which results in faster deci- sion-making in pressurized, inhospitable environments such as battlefields. The an- alytics enabled and performed opportunity the edge rep- resents. This has resulted in a widening gap between data capture and decision-mak- ing. “Based on the open architecture and the modularity, our intent here is to create an

pods off the queue, evaluates the pod’s requirements, and assigns it to a worker node. Placement Decision Stages: 1. Filter out impossible worker nodes a. Filters are called predicates - extensible in pods off the queue, evaluates the pod’s requirements, and assigns it to a worker node. Placement Decision Stages: 1. Filter out impossible worker nodes a. Filters are called predicates - extensible in pods off the queue, evaluates the pod’s requirements, and assigns it to a worker node. Placement Decision Stages: 1. Filter out impossible worker nodes a. Filters are called predicates - extensible in

pods off the queue, evaluates the pod’s requirements, and assigns it to a worker node. Placement Decision Stages: 1. Filter out impossible worker nodes a. Filters are called predicates - extensible in pods off the queue, evaluates the pod’s requirements, and assigns it to a worker node. Placement Decision Stages: 1. Filter out impossible worker nodes a. Filters are called predicates - extensible in pods off the queue, evaluates the pod’s requirements, and assigns it to a worker node. Placement Decision Stages: 1. Filter out impossible worker nodes a. Filters are called predicates - extensible in

Platform沉淀公共运维能力 • Operator Platform Kubernetes API Server Operator Manager sidecar framework 运维能力 operator sidecar framework 运维能力 operator 运维平台 运维基础 能力沉淀 运维平台 运维能力 编程框架 Kubernetes Platform不可变基础设施 base

Containers Demo: Serving the Model with TF Serving • Options for serving • Wrap model in a web framework (eg – Flask) • Tensorflow Serving • Seldon Demo: Run TensorFlow Training with Kubeflow

(AKS) 平台作为 Alauda 一系列平台的基础，可以与其他平台共同使用， 例如：Alauda Container Platform (ACP)，包括：DevOps、Service Framework、Service Alauda Kubernetes 白皮书 灵雀云 5 Mesh，Machine Learning 等，构成完整的产品闭环，可以根据企业定制化的使用场景，选

给Node打分 （软性选择） 绑定周期 • 向binding接⼝发送Post请求，把结果告诉k8s
 基于CPU实时使⽤率的调度器 借助Scheduling Framework可以实现⾃定义的调度器 score=(1 - cpu_used_seconds / cpu_total ) * Maxscore 未来规划 垂涎已久的架构 ServiceMesh 实际案例

倡导的“应用交付分层模型” OAM 项目近期计划 • OpenFaas、Terraform、Knative 集成 • K8s Operator 一键接入 • oamctl • oam-framework • CRD (traits/workloads) registry • … 主页：https://oam.dev 规范：https://github.com/oam-dev/spec

• Kubernetes DNS Resolver VS Kubernetes API Server Resolver • DNS resolver is builtin in gRPC framework and its out-of-box for users • When connection fail DNS Resolver can resolve name immediately •