Navigating the Compute Ocean: An Introduction to Kubernetes

Sujai Prakasam @TESCO

Agenda

• State of Infrastructure & Applications today

• Microservices and two pizza teams

• Orchestration challenge

• Containers and Docker

• Kubernetes to the rescue

• Kubernetes Architecture 101

• Demo

State of Infra & Apps today

• Application specific machine pools • Usually one application per machine • Inefficient and painful to manage at scale • Application are tightly coupled with the

underlying infrastructure • Patching / Configuration updates can potentially

cause application outages • Machines are really the primary key of datacenter

Microservices & 2 Pizza Teams

“Microservice architectural style is an approach to developing a single application as a suite of small services, each running in its own process and communicating with lightweight mechanisms, often an HTTP resource API. These services are built around business capabilities and independently deployable by fully automated deployment“ - Martin Fowler

“Small cross functional teams are responsible to build, test, deploy, monitor, scale and operate each of these services, teams communicate with each other via API contracts”

“In our conversations with microservice practitioners, we see a range of sizes of services. The largest sizes reported follow Amazon's notion of the Two Pizza Team (i.e. the whole team can be fed by two pizzas)” - Martin Fowler

Interactions between services running on Hailo’s platform.

Orchestration challenges

Linux Containers & Docker

• Linux processes not lightweight Virtual Machines

• Application + Dependencies = image

• Runtime environment (cgroups, namespaces, env vars)

• Docker made containers easy to build, share and run

Containers vs Virtual Machines

Kubernetes aka k8s

• Greek for “helmsman”; person who steers a ship

• Runs and manages containers at scale

• Inspired by Googles experience with running containers at scale

• Open Source & Supports almost all platforms

• IP owned by Cloud Native Computing Foundation

• Provides a Unified Compute Substrate

• Applications become the primary key of our datacenter / cloud infrastructure

Pod

• A pod is a group of one or more application containers which are relatively tightly coupled,

• Containers within a Pod are always co-located and co-scheduled, and run in a shared context

i.e. they share the same storage, IP address and port space, and can find each other via localhost

Useful Reading The Distributed System Tool Kit: Patterns for Composite Containers http://blog.kubernetes.io/2015/06/the-distributed-system-toolkit-patterns.html

Pod: Resource sharing and communication

Why not just run multiple programs in a single container? Decoupling software dependencies; The individual containers may be versioned, rebuilt and redeployed independently.

Labels & Selector

• Labels are key/value pairs that are attached to objects such as pods specify attributes meaningful and relevant to users

• Via a label selector, the client/user can identify a set of objects. The label selector is the core grouping primitive in Kubernetes.

Replication Controller

• A replication controller ensures that a specified number of pod “replicas” are running at any one time

• If there are too many pods, it will kill some. If there are too few, the replication controller will start more

• Pods maintained by a replication controller are automatically replaced if they fail, get deleted, or are terminated

Replication Controllers

Services

• Pods are mortals, they are automatically replaced by the replication controller if they fail, get deleted, or are terminated

• A Kubernetes Service is an abstraction which defines a logical set of Pods and a policy by which to access them

• Virtual IP per service

• Think of them as like internal load balancers with built-in service discovery

Key Concepts

….Off to Demo