Docker: The Future of Server Virtualization and Software Delivery

Docker (www.docker.com) is an open-source project that automates the deployment of applications inside software containers, by providing an additional layer of abstraction and automation of operating system–level virtualization on Linux.
Released barely more than 1 and half years ago, Docker is already in Production in numerous companies and in evaluation mode by countless others.

It is the most talked about project in the Cloud, Devops and Virtualization worlds in the past year and it has huge support from companies such as Google, Vmware, Redhat.

Introduction
- What is Docker?
- How do Docker Containers work?
Very Short Demo
Impact on
- Software Delivery
- Server Virtualization
Eco-system
Future Trends

Transcript

1. The Future of Server Virtualization and Software Delivery Presented by Steve Teo
2. Agenda Introduction What is Docker? How do Docker Containers work? Very Short Demo Impact on Software Delivery Server Virtualisation Eco-system Future Trends
3. Introduction “Everybody needs a grand entrance”
4. The Docker Project https://github.com/docker/docker Written in Golang Open source, Apache 2.0 license Current Release: 1.3.1 Current Project Activity 16307 11413 Commits 4989 Pull Reqs
5. The Docker Project Started as an internal project in dotCloud, a PAAS company Released on 13 March 2013 After pivoting, dotCloud was renamed Docker Inc. on 29 October 2013 PAAS business eventually sold off to fully focus on Docker http://blog.docker.com/2013/10/dotcloud- is-becoming-docker-inc/
6. Docker Inc. CEO: Solomon Hykes Value-generating activities: Docker-related products and services Creating a network of certified professional support, training, and services providers Funding: $15M Series B (Jan 2014), $40M Series C (Sept 2014) Currently Valuation: $400 million
7. Industry Support
8. Docker in Production The following from the back of my head Read about others @
 https://www.docker.com/resources/ usecases/
9. What is Docker?
10. Current Situation
11. Current Situation
12. The Solution
13. Docker Convenient CLI + Management around LXC Containers Operating system–level virtualization Runs only on Linux machines for now Kernel 3.8+ Container/Host Architecture must match On-going work for Docker Support on Windows Server/Microsoft Azure
14. Operating System-Level Virtualization? Virtual Machines Containers/OS-Level Vs
15. Core Components Docker Daemon: used to manage docker (LXC) containers on the host it runs Docker CLI: used to command and communicate with the docker daemon Docker image index: a repository (public or private) for docker images
 https://registry.hub.docker.com/
16. Core Elements Docker containers: directories containing everything-your- application Docker images: snapshots of containers or base OS (e.g. Ubuntu) images Dockerfiles: scripts automating the building process of images
17. Docker Containers Also known as Light-weight VMs Key Features Application portability Isolating processes Preventing tampering with host Managing resource consumption
18. Docker Containers Prevents Messing with other processes Causing "dependency hell” Or not working on a different system Being vulnerable to attacks and abuse all system's resources
19. Docker Architecture
20. Docker Workflow/Commands
21. How do Docker Containers work? “Down the Rabbit Hole”
22. LXC Heritage - FreeBSD Jails No device emulation Provides Namespacing of pid (process) net (networking) ipc (inter-process communication) mnt (mount) uts (hostname) Docker team has replaced LXC with lib container
23. Cgroups Linux Kernel feature Limit, account, and isolate resource usage of process groups CPU Memory Disk I/O
24. AUFS Another Union File System
25. AUFS Efficient Layering of Directory Structures Benefits Storage savings Faster Deployments Easier Upgrades Good Read http://blog.dotcloud.com/kernel- secrets-from-the-paas-garage-part-34-a
26. Want to know more? http://blog.dotcloud.com/paas- under-the-hood-ebook
27. Very Short Demo
28. Software Delivery “Sometimes, a radical idea might just solve the problem”
29. Build Once, Run Anywhere Avoiding dependency hell Each application comes with its own dependencies Eliminates dependency conflicts and extensive test cases Reduce/Eliminate concerns about compatibility between target platforms
30. Continuous Delivery Build containers which work from development all the way to production CI Platforms can be re-tuned to deploy containers for testing on different env Enables Immutable Infrastructure Fast delivery workflow, small changes encouraged instead of big bang changes
31. Continuous Delivery
32. Standardized Environments Easy for Sysadmins to set up standardised environments Ties in with configuration management tools Configuration management tools (eg. Chef, Puppet, Ansible) provide orchestration and bring host OS up to scratch (eg. install Docker) Docker takes over the application layer from that point onwards
33. Deploy and Scale More Easily Docker containers run (almost) everywhere Easy to move applications around Light-weight containers ensures greater scaling elasticity - easy to scale up and scale down
34. Enables thing not usually convenient Immutable Infrastructure - throwaway (easy to set up and tear down) Micro-services - one in every container?
35. Bridging the gap between Dev and Ops “Make Love, Not War”
36. Server Virtualisation “There is a ghost in every machine”
37. A look back in time The reasons for adoption of VMs Emulate hardware to run guest OS on differing host OS Utilize computing resources effectively Abstraction to resolve conflict contentions e.g. ports, cpu Easy provisioning of virtual machines (during the days of EC2)
38. VMs vs Container Virtual Machines Containers Vs
39. VMs vs Containers Containers setup and ‘boots’ in seconds compared to VMs Containers are more efficient when it comes to resources eg. space, ram Containers are isolated and can be limited as well, in its own way Qn: Do we really need the VM bacon then? Good reads http://www.infoq.com/news/2014/08/vm-containers- performance http://www.slideshare.net/Flux7Labs/performance-of- docker-vs-vms
40. Value and Cost-Savings Get higher density and run more workloads No need for hypervisors, pack more containers onto a host Get more value out of every server Reduce spending on equipment and licenses
41. The Ecosystem “Because nothing can survive alone”
42. CoreOS https://coreos.com/ Linux on Diet - Massive Server Deployments Light Footprint - 114MB Derived from ChromeOS Applications run as Docker Containers Painless, Safe Updates Clustered by Default and Service Discovery
43. Many more! Google Kubernetes (Container Management, Orchestration, Configuration)
 https://github.com/GoogleCloudPlatform/ kubernetes Redhat’s Project Atomic (Similar to the goals of CoreOS + Kubernetes)
 http://www.projectatomic.io http://www.mindmeister.com/389671722/ docker-ecosystem
44. Future Trends “The future might already be here”
45. Industry Support
46. Everybody embracing Docker Developers Sysadmins Data Centre Engineers Cloud Engineers CTOs/CIOs/Project Managers
47. ~2000 2014 Long Lived Development is iterative and constant Monolithic and built on a single stack Built from loosely coupled components Deployed to a single server Deployed to a multitude of servers Shift in Application Delivery
48. Shift in Architecture
49. Essential Job Skill Core technology for companies Core skill which recruiters would look out for in resumes
50. Changes to the Cloud Status Quo Shift in IAAS providers to move away from virtualization towards bare metal servers
 http://virtualizationreview.com/articles/ 2014/10/29/containers-virtual-machines- and-docker.aspx Most Public PAAS will support Docker as a first-class citizen OpenStack + Openshift + Docker = Marriage made in heaven, enables private PAAS for enterprises
51. Before I End Imagine you had the chance to adopt Docker, ask yourself some questions What kind of use-cases can you come up for leveraging Docker? Do our servers really need virtual machines or can we actually go bare metal? How to manage and orchestrate massive docker scaling?
52. Community Docker Singapore Meetup http://www.meetup.com/Docker- Singapore/ Docker Community Page https://www.docker.com/ community/
53. Thank You Any Questions? M: Steve Teo E: xylancer@gmail.com T: @kelltainer