LinuxCon ContainerCon CloudOpen China

Kubernetes currently has two load balancing mode: userspace and IPTables. They both have limitation on scalability and performance. We introduced IPVS as third kube-proxy mode which scales kubernetes load balancer to support 50,000 services. Beyond that, control plane needs to be optimized in order to deploy 50,000 services. We will introduce alternative solutions and our prototypes with detailed performance data.

Resource placement is a policy-rich problem, particularly across multi-cluster, multi-geography and multi-cloud environments. Placement may be based on company conventions, external regulation, pricing, performance requirements, or complex combinations of those. Furthermore, placement policies evolve over time and vary across organizations. As a result, it is very difficult to anticipate the policy requirements of all users.

In this presentation, Torin Sandal (Lead Engineer of Open Policy Agent) will present, along with Irfan Ur Rehman, and demonstrate the work they've done integrating OPA into the Kubernetes Cluster Federation Control Plane. This enables high level policies to be expressed in a easy to understand policy language, and automatically enforced across federations of Kubernetes clusters.

Kubernetes is a powerful open source system, and how to evolve it into tools or production scenario is a hot topic. In this talk we'll present how Huawei builds a global distributed IT system for application development, resource management, and scheduling with Kubernetes, and how it reduces the application deployment from one week to 10 minutes. We will also talk about our enhancements on monitoring, logging, security, volume, DFX, etc, which make Kubernetes production ready and support Huawei IT application migration to Cloud Native. Furthermore, we will introduce the problems we met and our solutions.

Fully Automated Kubernetes Deployment and Management (Peng Jiang, Rancher Labs) - Kubernetes is rapidly gaining popularity as a powerful container orchestration and scheduling platform. But deploying and managing Kubernetes clusters is still a challenge for many organizations.How to ensure Kubernetes clusters in different clouds and data centers can communicate with each other? How to automate the deployment of multiple Kubernetes clusters? How to incorporate the new Kubernetes Federation into multi cloud and multi datacenter deployments? How to manage the health of Kubernetes cluster itself? etc.

In this talk, Peng will share his experience on how to automate and simplify Kubernetes deployments, and discuss how some of the latest community projects (such as kubeadm and self-hosting Kubernetes) will help address the problems in the future.

A Major breakthrough in cloud-native technology is driving the giant telecom and cable industries to transition from the existing VM based infrastructure to Container Orchestration Engines like Kubernetes, which has given rise to the need to support multi-interface pods. Isolation of network traffic into different network planes such as management, control and media is an indispensable feature of telecom infrastructure. Kubernetes doesn’t have support for multi-network and network isolation, and scheduling criteria for pods currently only to takes compute resources into consideration and is agnostic to other opaque resources like SRIOV. In the presentation, Kural and Swati demonstrate a dynamic solution for advanced network scheduling and in order to take advantage of node’s hardware and network capabilities using Multus & SRIOV CNI and Node feature discovery in Kubernetes.

Storage is a critical part of running containers, and Kubernetes offers some powerful primitives for managing it. This talk will cover a variety of volume plugins supported in Kubernetes and their current state and usage. Then it will focus on the major components in the storage system to enable automatic attaching/detaching network attached volumes, dynamic provisioning, and storage capacity management.

Effective management of shared resources (CPU, memory, etc.) is critical to achieving high utilization and strong isolation in shared clusters. This talk will describe the Kubernetes features that allow you to simultaneously achieve your utilization, isolation, and availability goals. We will also examine the new features in Kubernetes 1.6 that give you fine-grained control over placement of your containers, and describe how to use them to achieve various concrete policy goals. Lastly we will discuss the feature roadmap in this area for the next few releases.

Cloud Native architectures, like those built on Kubernetes, are envied for their ability to segment different parts of an application into microservices to be separately containerized, so that each microservice can be written in its own language, with its own framework & unique set of libraries. This has been shown to significantly increase developer & team productivity by decoupling dependencies between different parts of a team. However, most legacy code consists of a large monolithic code base, not microservices applications. In this talk, Dan Kohn, executive director of CNCF, will walk through how monoliths evolve into modern cloud native architectures by taking advantage of the DevOps revolution & result in developer productivity gains.

The Kubernetes API provides new ways to build automation tools that are highly available and scale to serve clusters with many thousands of nodes. During this session attendees will learn how to move beyond shell scripts and configuration management tools and leverage cluster level APIs and distributed systems design patterns to build next generation automation tools. Attendees will learn how to extend the Kubernetes API and utilize their own custom declarative configuration to build controllers that automate job scheduling or schedulers that allow you to place jobs based on your rules.

We have witnessed the soaring interest in Kubernetes among the Chinese developers, and how Kubernetes has empowered even the traditional, state-owned enterprises to reform their legacy production systems to excel in the “Internet+ transformation” wave in China. 

We’ll share the tale of how Kubernetes has evolved to win the container cluster management battle in China, with unprecedented interviews and stats from both the local developer community and Chinese Fortune 500 enterprises. Looking forward, we also shed lights on how to better localize Kubernetes in China and thus propel its adoption globally.