Getting Started with Cloud Native

#1 Cloud Native Surat Meetup Slides

Transcript

1. Getting Started with Cloud Native Cloud Native Surat Meetup 16 July, 2017
2. Agenda • Introduction about Cloud Native Computing Foundation • Brief overview about projects under CNCF • How to get started, integrate projects in your application • Q/A, discussion, knowledge sharing regarding projects • Open discussion about next meetup & topic
3. What is Cloud Computing?
4. “If someone asks me what cloud computing is, I try not to get bogged down with definitions. I tell them that, simply put, cloud computing is a better way to run your business.” ~ Marc Benioff, Founder, CEO and Chairman of Salesforce
5. 1. Fast 2. Secure 3. Global scale 4. Cost-effective 5. Reliability Cloud Computing
6. What is Cloud Native Computing Foundation (CNCF)? • CNCF is a nonprofit organization committed to advancing the development of cloud native applications and services • Announced on 21 July, 2015 • Under Linux Foundation umbrella • All projects are open-source
7. Software and patterns for... Container based Microservices oriented Programmable infrastructure Cloud Native
8. • Initiative to spread awareness about cloud development • Help/Guide developers on how to work on cloud environment • Promote developers to contribute in open source projects • Discuss about latest cloud native projects • Share meetup members cloud development experience with others Cloud Native Surat Meetup
9. Meetup Effect
10. Kubernetes Fluentd Prometheus OpenTracing Linkerd CNCF Hosted Projects
11. gRPC rkt CoreDNS containerd CNI CNCF Hosted Projects
13. Kubernetes • Automating deployment, scaling, and management of containerized applications • Developed by Google • Highly inspired from Google’s internal tool “Borg” • Type: Container Orchestration • Project accepted: March 10, 2016 • Similar projects: ◦ Docker Swarm, Mesos, Nomad
14. • Automatic binpacking • Automated rollouts and rollbacks • Service discovery and load balancing • Storage orchestration • Secret and configuration management • Self-healing • Horizontal scaling • Batch execution Kubernetes Features
15. • Monitoring system and alerting toolkit • Helps with metrics and monitoring • Started in SoundCloud • Type: Monitoring • Project accepted: April 9, 2016 • Similar projects: ◦ Graphite, InfluxDB, Nagios Prometheus
16. • Dimensional data • Powerful queries • Great visualization • Efficient Storage • Precise alerting Prometheus Features
17. • Provides insight about individual requests as they propagate throughout a system • Focus on event tracing not event logging • consider all requests from all services, then connect the dots • Type: Tracing • Project accepted: October 11, 2016 • Use case: ◦ Application Code, Library Code, OSS Services, RPC/IPC Frameworks OpenTracing
18. OpenTracing
19. • Detailed logs for any component • Request counts / Operation counts • Distributions of latency measurements • Peer relationships within distributed systems ◦ … and thus service dependency graphs Why use OpenTracing?
20. • Lightweight Unified Logging Layer • Invented by Treasure Data • Type: Logging • Project accepted: November 9, 2016 • Similar projects: ◦ Logstash, Apache Flume • Use Case: ◦ Centralized App Logging, Log Management & Search, Data Archiving Fluentd
21. Before Fluentd
22. After Fluentd
23. • Pluggable Architecture • More than 600 plugins available • Full integration with Docker and Kubernetes • Libraries in many languages Fluentd Features
24. • Works as a proxy • layer that manages the communication between apps or parts of the same app • Developed by buoyant • Type: Service Mesh • Similar projects: ◦ Envoy, Istio • Project accepted: January 23, 2017 Linkerd
25. • HTTP proxy integration • Runtime traffic routing • Latency-aware load-balancing • Distributed tracing • Protocol upgrades • Drop-in service discovery Linkerd Features
26. • RPC framework • Use Protocol Buffers for serializing structured data • enables client and server applications to communicate • Type: Remote Procedure Call • Developed by Google • Project accepted: March 1, 2017 • Similar project: ◦ Apache Thrift gRPC
27. gRPC
28. • Low latency, highly scalable, distributed systems • Developing mobile clients which are communicating to a cloud server • Bi-directional streaming and integrated auth • HTTP/2 • Layered design for extensions gRPC Features
29. • Distributed service for announcement and discovery of services • Utilizes DNS queries to discover available services • Server framework developed as part of the web server Caddy • Can run in place of the standard Kube-DNS in Kubernetes • Type: Service Discovery • Project accepted: March 2, 2017 CoreDNS
30. • Load balancing of responses • Automatically load zone files from disk • Health checking endpoint • Provide metrics (by using Prometheus) • Caching • Serve as a proxy to forward queries to some other nameserver • Middleware support CoreDNS Features
31. • Based on the Docker Engine’s core container runtime • It contains two specifications: ◦ runtime-spec: The runtime specification ◦ image-spec: The image specification • Developed by Docker • Type: Container Runtime • Project accepted: March 29, 2017 containerd
32. containerd
33. • OCI(Open Container Initiative) Image Spec support • Container execution and supervision • OCI Runtime Spec support (aka runC) • Network Interfaces Management • Local storage • Native plumbing level API containerd Features
34. • CLI for running application containers on Linux • automating deployment, scaling, and management of containerized applications • Developed by CoreOS • Type: Container Engine • Similar projects: ◦ Docker • Project accepted: March 29, 2017 rkt(rocket)
35. • Pod-native • Security • Composability • Open standards and compatibility rkt Features
36. • Goal is to create a generic plugin-based networking solution for containers • Kubernetes, rkt natively supports CNI • Proposed by CoreOS • Type: Networking • Project accepted: March 29, 2017 CNI(Container Network Interface)
38. • Pluggable networking interface • Reduce duplicate work for container runtimes and orchestrators • Common interface between the network plugins and container execution Why CNI?
39. Questions
40. How to get started?
41. Show-off your SKILLS
42. • Fix small errors(typos, broken links) • Read issues, answer them if you know • Don’t hesitate for opening new issue • Try for small pull requests • Improve sample code Contribute in open source projects
43. Helpful links CNCF Projects (https://www.cncf.io/projects) TOC meeting (https://lists.cncf.io/mailman/listinfo/cncf-toc) The New Stack (https://thenewstack.io) Google Cloud Platform Podcast (https://www.gcppodcast.com) Kube Weekly(https://kubeweekly.com/introducing-kubeweekly/) Docker (https://www.docker.com)
44. Thanks! Yash Thakkar https://github.com/thakkaryash94 @mrviyet