Docker Desktop is the easiest - and most popular - way for developers to get started with Docker and Kubernetes. It’s a tool that is designed for both beginners and advanced Docker users and we’ve added a number of enhancements over the last 6 months to make it easier for developers to “shift left” and deliver production-ready applications. We’ll demonstrate how application templates are built and used and also how Docker Desktop integrates with your favorite development tools. You'll see how easy it can be for developers that are new to Docker to get started with container-based development. If you're a Docker pro or an application architect we'll show you how to Docker Desktop enables you to control the application and container specs. Finally, we’ll also highlight some of the things we’re working on to further enhance the desktop experience.
3. ● Late stage bugs
● Manual integration flows
● Learning new syntax and tools
Developers want to ship clean code, fast
What slows you down?
4. ● Predictable, isolated environments
● Include application dependencies
● Same functionality across
environments
● Ship new code faster
Containers provide
consistency...
5. • Build images with Dockerfiles
• Architect applications for Swarm
and Kubernetes
...And introduce new
complexity
7. Introducing Docker Desktop
Reduce Developer Friction Dev to Prod - Faster
Build Containerized Apps
with Ease
Shareable and Reusable
Multi-Service Apps
8. Reduce Developer Friction
Containers run natively and securely
on the desktop
● Remove the need for virtual machine
development
● Run various application types natively
and concurrently
● Pre-set administrative settings
● Simple roll out to development teams
○ Windows MSI and Mac PKG files
11. Accelerate Time to Production
TEST STAGING
Scanning SigningAutomated Policies
PRODUCTION
● Version parity from Dev to Prod
● Roll out new clusters with confidence
kubernetes 1.14
docker API v1.39
kubernetes 1.14
docker API v1.39
12. •Desktop GUI to build new
applications faster
•Desktop provides ready-to-run
Dockerfiles and Compose Files
Build Containerized
Apps Faster
14. ● Templated applications make apps
easier to implement and reuse
● Template-based workflows for the
entire organization
Make multi-service
apps more useable
and shareable
16. What’s Next: Pipelines
Bootstrap CI/CD pipelines for Docker
KEY FEATURES
- Usable via CLI, IDE or GUI
- Work with CNAB bundles or Dockerfile
- Integrate with third-party CI systems like Jenkins
- Reduce friction to using Docker Content Trust
- Focused on bootstrapping projects quickly
$ docker pipeline create
Which build system do you want
to use? jenkins
Which Git repository is the
code stored in?
17. What’s Next: Buildx for Arm
Build Arm images from your x86 Desktop
KEY FEATURES
- Extends on Build functionality
- No changes needed to existing Dockerfiles to
support Arm processors
- Build multi-arch Arm images locally on x86
development machines using Docker for Desktop
- Share and collaborate through Docker Hub
Sign up for Tech Preview at https://beta.docker.com
18. Reduce Developer Friction Dev to Prod - Faster
Build Containerized Apps
with Ease
Shareable and Reusable
Multi-Service Apps
Docker Desktop Driving Developer Efficiency
19. Nick Woodbridge | GlaxoSmithKline• Nick Woodbridge
• 15 years experience building and selling products across Financial Services, Cloud (AWS and
Adobe), and now Pharma
• Lead 4 product development teams focusing on data ingestion, data management, and data
visualization
• GSK – 300 year old Pharma Company
• New strategy to push heavily into genetics
• Hadoop as the data lake
• PBs of information
• Organization is Data and Computational Sciences
• Work with DS and AI / ML teams to productionalize their algorithms and visualizations
• Using Docker EE for management of custom developed products as well as COTS solution
• Engineering teams delivering with Docker Desktop bespoke solutions
Sign up for Tech Preview at https://beta.docker.com
20. Sign up for Tech Preview at https://beta.docker.com
Environment – Data Science Teams
Data Lake
Exploratory Data Science – Using
Docker EE / Kubernetes
1) ENOD (Edge Node on
Demand)
2) Computational Notebooks
Production - Cluster Docker EE
(K8s)
1) Productionalized Solutions –
using Algorithms created in
exploratory side
2) COTS Solutions
No Easy
Way
through
Here
21. Sign up for Tech Preview at https://beta.docker.com
Environment Future – Data Science
Teams
Data Lake
Exploratory Data Science – Using
Docker EE / Kubernetes
1) ENOD (Edge Node on
Demand)
2) Computational Notebooks
Production - Cluster Docker EE
(K8s)
1) Productionalized Solutions –
using Algorithms created in
exploratory side
2) COTS Solutions
Docker Desktop