Slide 1

Slide 1 text

No content

Slide 2

Slide 2 text

No content

Slide 3

Slide 3 text

Navigating Google Cloud Platform Your first week on GCP +Terrence Ryan @tpryan

Slide 4

Slide 4 text

@tpryan Introduction

Slide 5

Slide 5 text

@tpryan Who are you?

Slide 6

Slide 6 text

@tpryan Help developers and sysadmins new to GCP identify the stuff they need Goal

Slide 7

Slide 7 text

@tpryan

Slide 8

Slide 8 text

@tpryan (╯°□°)╯︵ ┻━┻

Slide 9

Slide 9 text

@tpryan ● Computing ● Networking ● Storage ● Big data ● Administration ● Development ● Machine learning Agenda

Slide 10

Slide 10 text

@tpryan Computing Doing things

Slide 11

Slide 11 text

@tpryan Virtual machines Containers Serverless Computing

Slide 12

Slide 12 text

@tpryan Virtual Machines ● 1 - 64 processors ● .6 - 416 GB (.9 - 6.5 GB per proc) ● Disk limits ­ 65 TB Persistent ­ 3 TB SSD ­ 208 RAM ● Preconfigured images ● Can build your own ● Spin up in 10s of seconds Compute Engine

Slide 13

Slide 13 text

@tpryan Virtual Machines ● Prebuilt software images ● Single systems ● Groups of systems Cloud Launcher

Slide 14

Slide 14 text

@tpryan Compute Engine Demo Kick the tires and launch a few VMs

Slide 15

Slide 15 text

@tpryan Containers

Slide 16

Slide 16 text

@tpryan Containers ● Managed Kubernetes Container Engine

Slide 17

Slide 17 text

@tpryan Kuber-what-es?

Slide 18

Slide 18 text

@tpryan Frontend 10.0.0.1 Public 130.91.x.x1 API 10.0.0.2 Public 130.91.x.x2 Data 10.0.0.3 Kubernetes ● Container Orchestration System ● Allows for running a large number of containers in production ● Organizing them into workloads ­ Microservices Scheduled jobs Batch jobs

Slide 19

Slide 19 text

@tpryan DNS Server Logging Service Discovery Monitoring ! Kubernetes

Slide 20

Slide 20 text

@tpryan DNS Server Logging Service Discovery Monitoring !

Slide 21

Slide 21 text

@tpryan Container Engine demo Launch a few services

Slide 22

Slide 22 text

@tpryan ● Managed Kubernetes (Containers as a service) ● Smart defaults for ­ Monitoring ­ Logging ­ DNS ­ Service Discovery ● Auto scaling Container Engine ● Custom Runtime ­ Bring your own Dockerfile ● Autoscaling ● Given one endpoint per service App Engine flexible environment

Slide 23

Slide 23 text

@tpryan ● Container image storage ● Works with Container Engine App Engine flexible Docker Container Registry ● Container building as service ● Integrates with Container Registry ● Integrated with build processes Container Builder

Slide 24

Slide 24 text

@tpryan ● Languages ­ Java 8 / Servlet 3.1 / ­ Jetty 9 ­ Python 2.7 & 3.5 ­ Node.js ­ Ruby ­ PHP ­ Go ­ Open capabilities App Engine flexible runtime ● Languages ­ Python 2.7 ­ Java 7 ­ PHP 5.5 ­ Go ● Constrained capabilities ● Incredibly rapid scale App Engine standard runtime ● Javascript / Node.js ● Respond to events Cloud functions

Slide 25

Slide 25 text

@tpryan App Engine Load Demo Quick demo showing App Engine in comparison to Compute Engine

Slide 26

Slide 26 text

@tpryan Networking Connecting things

Slide 27

Slide 27 text

@tpryan ● Create custom private cloud networks ● Set up custom routes to your other non-Google locations Cloud Virtual Network ● Create a dedicated connection to Google ● Direct Peer to one of our local or regional Points of Presence Cloud Interconnect

Slide 28

Slide 28 text

@tpryan ● Gateway to Gateway Only ● Can be used to connect to any IPSEC compliant gateway VPN ● Set up rules ● Apply rules to tags ● Apply tags to hosts Firewall

Slide 29

Slide 29 text

@tpryan ● Activate through our load balancers ● Uses HTTP headers to determine cacheability CDN ● Manage DNS on Google’s network ● SLA promises 100% availability of our Authoritative Name Servers DNS ● HTTP ● SSL ● Internal ● Network Load Balancers

Slide 30

Slide 30 text

@tpryan What if I need something you don’t offer?

Slide 31

Slide 31 text

@tpryan ● Build by hand Compute Engine ● Prebuilt solutions for: ­ Barracuda ­ Brocade ­ Cloudflare ­ more... Cloud Launcher

Slide 32

Slide 32 text

@tpryan Storage Keeping things

Slide 33

Slide 33 text

@tpryan Files Databases Big Data Storage

Slide 34

Slide 34 text

@tpryan File Storage ● One set of interfaces ● Several types ­ Multi-regional ­ Regional ­ Nearline ­ Coldline Cloud Storage

Slide 35

Slide 35 text

@tpryan Public Content Processed Data Long Term Storage File Storage

Slide 36

Slide 36 text

@tpryan Public Processed Long Term Storage Characteristics 99.95% availability Geo-redundant 99.9% availability Data stored in a narrow geographic region 99.0% availability 30-day minimum storage duration 99.0% availability 90-day minimum storage duration Uses Cases Frequently accessed World Audience Website content Streaming video Gaming Mobile Accessed by your system But not by your customers Big Data sets for processing Backup Long tail multimedia Disaster recovery Compliance archives Storage Cost GB/month $.026 $.02 $.01 $.007 Retrieval Cost GB $.01 $.05 Storage Type Multi-region Regional Nearline Coldline

Slide 37

Slide 37 text

@tpryan NoSQL Application data SQL

Slide 38

Slide 38 text

@tpryan ● Traditional SQL Server ● Vertically scalable ● Managed backups ● Easy to setup replicas ● Types ­ MySQL ­ Postgres Cloud SQL ● Structured data ● Highly available ● Strongly consistent ● CAP theorem says impossible ● Globally distributed ● Horizontally scalable Cloud Spanner

Slide 39

Slide 39 text

@tpryan ● Document based ● Indexable ● Giant Cloud Datastore ● Columnar ● Low latency ● Hbase API compatible ● Needs scale to be cost effective Cloud Bigtable

Slide 40

Slide 40 text

@tpryan Databases Big data Storage Files

Slide 41

Slide 41 text

@tpryan Where do you store Big Data?

Slide 42

Slide 42 text

@tpryan Cloud Datastore Cloud SQL Cloud Spanner Cloud Bigtable Cloud Storage

Slide 43

Slide 43 text

@tpryan Big Data ● Pipe in semi-structured data ● Analyze it quickly ● Use SQL to do it. BigQuery

Slide 44

Slide 44 text

@tpryan BigQuery Demo Quick demo showing BigQuery

Slide 45

Slide 45 text

@tpryan ● Messaging Bus ● Many to many Pub/Sub ● Managed service for data analysis ● Apache Beam Dataflow ● Managed Spark and Hadoop ● With Autoscaling DataProc

Slide 46

Slide 46 text

@tpryan Cloud Datastore Cloud SQL Cloud Spanner Cloud Bigtable BigQuery Translation API Natural Language API Speech API Vision API Cloud Dataflow Compute Engine App Engine Container Engine Cloud Functions Cloud Pub/Sub Cloud Storage

Slide 47

Slide 47 text

@tpryan What if I need something you don’t offer?

Slide 48

Slide 48 text

@tpryan ● Build by hand Compute Engine ● Prebuilt solutions for: ­ Cassandra Redis Mongo CouchDB more... Cloud Launcher

Slide 49

Slide 49 text

@tpryan Administration Operations

Slide 50

Slide 50 text

@tpryan ● Cross Cloud Logging ­ GCP AWS On Prem possible Stackdriver Logging ● Cross Cloud Monitoring ­ GCP ­ AWS ● Alerts based off ­ Logs ­ Uptime checks ­ Built in or Custom Metrics Stackdriver Monitoring

Slide 51

Slide 51 text

@tpryan Security ● Allow for group of resources ● Role based permissions ● Fine grained controls ● Vary from product to product Cloud IAM

Slide 52

Slide 52 text

@tpryan ● Gcloud ● gsutil CLI ● SDK for most languages ● REST endpoints ● Functional equivalent to gcloud API ● Repeatable Complex deployments ● Configs ● Templates ● Disk Images ● Network setup ● more Deployment Manager

Slide 53

Slide 53 text

@tpryan Development Code

Slide 54

Slide 54 text

@tpryan ● Gcloud ● gsutil SDK ● IntelliJ ● Android Studio ● Powershell ● Visual Studio ● Eclipse Plugins ● Cloud Source Repository Others

Slide 55

Slide 55 text

@tpryan Reporting ● Automatic ­ App Engine ­ Load Balancers ● Enabled through an SDK ­ Java ­ Node.js ­ Go ­ Ruby ­ Zipkin ● Links to logs Stackdriver Trace

Slide 56

Slide 56 text

@tpryan

Slide 57

Slide 57 text

@tpryan Reporting ● Aggregate ● Logs or API ● Alerts Stackdriver Error Reporting

Slide 58

Slide 58 text

@tpryan

Slide 59

Slide 59 text

@tpryan Debugging ● App Engine Java Python Node.js Ruby ● Compute Engine Java Python Go Stackdriver Debugging

Slide 60

Slide 60 text

@tpryan Services API Manager

Slide 61

Slide 61 text

@tpryan Using Models Creating Models Extending Models Machine Learning

Slide 62

Slide 62 text

@tpryan Natural Language API ● Syntax Analysis ● Entity Recognition ● Sentiment Analysis ● Multi-Language Speech API ● Automatic Speech Recognition ● Streaming ● 80 languages Translation API ● Language Detection ● Translation ● Whole document Vision API ● Face Detection ● Label Detection ● Logo Detection ● OCR Machine Learning

Slide 63

Slide 63 text

@tpryan Machine Learning ● Managed Tensor Flow Cloud ML Engine

Slide 64

Slide 64 text

@tpryan What-flow?

Slide 65

Slide 65 text

@tpryan Machine Learning ● Open Source ML Library ● Builds models ● Allows you to compute with variable levels of CPU/GPU power Tensor Flow

Slide 66

Slide 66 text

@tpryan Machine Learning ● Managed Tensor Flow ● Self-tuning ● Portable Cloud ML Engine

Slide 67

Slide 67 text

@tpryan Conclusions ┬─┬ ノ( ゜-゜ノ)

Slide 68

Slide 68 text

@tpryan There’s a lot of stuff here

Slide 69

Slide 69 text

@tpryan

Slide 70

Slide 70 text

@tpryan What’s next? Start a free trial cloud.google.com/free Stop by Dome Sandbox H

Slide 71

Slide 71 text

Thank you! +Terrence Ryan @tpryan

Slide 72

Slide 72 text

Thank you @tpryan

Slide 73

Slide 73 text

@tpryan Connection Security Availability Networking

Slide 74

Slide 74 text

@tpryan Reporting Security Configuration Administration

Slide 75

Slide 75 text

@tpryan Coding Reporting Services Development

Slide 76

Slide 76 text

@tpryan Development