Book Goog Cloudonboard Northam v2

<CloudOnBoard>
Getting Started
With Google Cloud
</CloudOnBoard>
Cloud OnBoard
Welcome to Cloud OnBoard

#GoogleCloudOnBoard
©Google Inc. or its affiliates. All rights reserved. Do not distribute.

May only be taught by Google Cloud Platform Authorized Trainers.
Cloud OnBoard

Page 1
Agenda Why choose Google Cloud Platform?
Google Cloud Platform enables developers to build, test and

1 Introduction to Google Cloud Platform deploy applications on Google’s highly-scalable, secure, and
reliable infrastructure.
2 Quiz
Choose from computing, storage, big data/machine learning,
and application services for your web, mobile, analytics, and
backend solutions.
©Google Inc. or its affiliates. All rights reserved. Do not distribute. 3 ©Google Inc. or its affiliates. All rights reserved. Do not distribute. 5
The Future of Cloud Computing GCP is organized into regions and zones
Now Next
● Regions: collections of zones
○ Specific geographical locations where you can run resources
○ Regions are interconnected using Google’s global, meshed
backbone network
Storage Processing Memory Network Storage Processing Memory Network
● Zones: isolated deployment areas in a region
Physical/Colo Virtualized Serverless/No-Ops ● Your resources can be regional, zonal, or in some cases
multi-regional
User-configured, managed, and maintained Fully automated
Page 2 Page 3
Google’s infrastructure: the most powerful on earth Google’s commitment to
environmental responsibility
Developing our infrastructure while respecting
Belgium 2 Finland
London 3 Tokyo (2016)
Oregon 2 3
3
Iowa 4
3
3 N Virginia
Frankfurt
3 our ecosystem
S Carolina Taiwan
3
3 Mumbai
• Pioneering data center efficiency

2 Singapore
• Largest private investor in renewables (wind, solar)
Network path 3 São Paulo
Sydney
• First data centers to receive ISO 14001 certification
3
#
Current regions
and number of zones • 100% carbon neutral since 2007
Committed regions for 2017
# and number of zones
https://peering.google.com
https://cloud.google.com/compute/docs/regions-zones/regions-zones Image by Connie Zhou
©Google Inc. or its affiliates. All rights reserved. Do not distribute. 9
Google’s network: Innovative, Customer-Friendly Pricing

the world’s largest software-defined network
Belgium 2 Finland ● Billing in sub-hour increments
London 3 Tokyo (2016)
Oregon 2 3
3
Iowa 4
3
3 N Virginia
Frankfurt
3 ● Discounts for sustained-use
S Carolina Taiwan
3
○ Automatically reward users who run virtual machines for
3 Mumbai
2 Singapore over 25% of any calendar month

Network path
Points of presence
3 São Paulo
Sydney
● Custom virtual-machine types
3
Current regions
○ Pay only for the resources you need for your application
Committed regions for 2017

https://peering.google.com
https://cloud.google.com/compute/docs/regions-zones/regions-zones Image by Connie Zhou
Page 4 Page 5
Google Cloud Platform
Google’s commitment to open APIs and open source
Compute Storage
Container Cloud Bigtable Cloud Cloud SQL Cloud Cloud

Compute App Engine
Engine Functions Storage Spanner Datastore
Engine Beta
Beta
Big Data
Machine Learning
Customers should use us because they love us, Natural Machine Speech Translate API
BigQuery Pub/Sub Dataflow Dataproc Datalab Vision API
not because they are unable to migrate Language API Learning API
GCP offers a range of compute architectures
Stackdriver helps engineers keep their

cloud-powered applications fast and
available.
Compute Container App Cloud Managed

Engine Engine Engine Functions
Beta services Monitoring Logging Debug
IaaS Hybrid PaaS Serverless Automated

logic elastic
resources
Error Reporting Trace
Toward managed infrastructure Toward dynamic infrastructure
©Google Inc. or its affiliates. All rights reserved. Do not distribute. 12 ©Google Inc. or its affiliates. All rights reserved. Do not distribute.
Page 6 Page 7
Agenda Resources
● Why Google Cloud Platform?
https://cloud.google.com/why-google/
1 Introduction to Google Cloud Platform
● Pricing philosophy
https://cloud.google.com/pricing/philosophy/
2 Quiz
● Data Centers
https://www.google.com/about/datacenters/
● Google Cloud Platform product overview
http://cloud.google.com/products/
● Google Cloud Platform solutions
http://cloud.google.com/solutions/
Cloud OnBoard
Quiz
Name 3 pricing innovations that make Google Cloud Platform the
most cost-effective public cloud.
Sub-hour billing, sustained-use discounts, Compute Engine

custom machine types
In addition to innovative pricing, name 3 benefits of using Google

Cloud Platform.
Commitment to environmental responsibility, commitment to

open source technologies, robust infrastructure

Page 8 Page 9
Agenda Projects are uniquely named
● Projects use three identifying attributes:

1 Google Cloud Platform Projects
○ Project Name
○ Project Number
2 Identity and Access Management (IAM) ○ Project ID
■ Also known as Application ID
3 Interacting with Google Cloud Platform
● Interact with projects using the Cloud Console
4 Quiz or the Cloud Resource Manager API
Projects organize resources Agenda

● All Google Cloud Platform services
are associated with a project that is 1 Google Cloud Platform Projects
used to:
○ Track resource and quota usage
2 Identity and Access Management (IAM)
○ Enable billing
○ Manage permissions and credentials 3 Interacting with Google Cloud Platform
○ Enable services and APIs
4 Quiz & Lab
Page 10 Page 11
Identity and Access Management You can define “curated roles”
List of Permissions
Google
Group ✔ compute.instances.delete
✔ compute.instances.get
InstanceAdmin ✔ compute.instances.list
Role ✔ compute.instances.setMachineType
Who can do what on which resource ✔ compute.instances.start
✔ compute.instances.stop
project_a ...
IAM predefines “primitive roles” IAM Resource Hierarchy

● A policy is set on a
Organization
resource
example.com
○ Each policy
contains: Set of
Policy Inheritance
roles, role members
Project
Owner Editor Viewer Billing administrator ● Resources inherit bookshelf static-assets stream-ingest
policies from parent
Invite members Deploy applications Read-only Manage billing ○ Resource policies
Remove members Modify code access Add administrators are a union of
Can delete project Configure services Remove parent and resource Compute App Cloud Cloud Cloud BigQuery
Includes Editor rights Includes Viewer rights administrators
Resources
Engine Engine Storage Storage Pub/Sub
● If parent policy less
restrictive, overrides
more restrictive
A project can have multiple owners, editors, viewers and billing administrators. resource policy
instance_a queue_a bucket_a bucket_b topic_a dataset_a
Page 12 Page 13
Organization node organizes projects Service Accounts and IAM
● Organization node is root node Visibility ● Service accounts
Control authenticate with keys Identity IAM Role Resource
for Google Cloud resources
○ Can be managed via Cloud ○ Google manages keys, key
Resource Manager API rotation for Compute Engine
bob@example.com example.com and App Engine
Service Account InstanceAdmin Role Compute Instances
Organization
● 2 organization roles: ● Can assign an IAM role to the
Admin
○ Organization Admin - Control service account
Create
over all cloud resources ● Can also assign
○ Project Creator - Controls ServiceAccountActor role
alice@example.com project_1 project_2 User/Group ServiceAccountActor Role Service Account
project creation to users/groups
Project Creator
Service Accounts Example: Service Accounts and IAM

project_a project_b
● Provide an identity for carrying out server-to-server ● VMs running component_1 are
granted Editor access to
interactions in a project
project_b using Service Account 1
Service
● Used to authenticate from one service to another
component_1
Account 1
● VMs running component_2 are Editor
granted objectViewer access to

● Can be used with primitive and curated roles
bucket_1 using Service Account 2
● Identified with an email address: ● Service account permissions can component_2 Service
Account 2
PROJECT_NUMBER@developer.gserviceaccount.com be changed without recreating Storage.
objectViewer
PROJECT_ID@developer.gserviceaccount.com VMs
bucket_1
Page 14 Page 15
Agenda Google Cloud Platform Console
● Centralized console for all project data
1 Google Cloud Platform Projects ● Developer tools
○ Cloud Source Repositories
2 Identity and Access Management (IAM)
○ Cloud Shell
○ Test Lab (mobile app testing)
3 Interacting with Google Cloud Platform
● Access to product APIs
4 Quiz ● Manage, create projects
Interacting with Google Cloud Platform Google Cloud SDK
Cloud SDK/
● SDK includes CLI tools for Cloud
Cloud Console >_ Cloud Shell Platform products and services
Web user interface
command-line interfaces
○ gcloud, gsutil (Cloud Storage), bq

(BigQuery)
● Available as Docker image

● Available via Cloud Shell
Cloud Console
REST-based API Mobile App ○ Containerized version of Cloud SDK running
For Android and iOS
on Compute Engine instance
32
©Google Inc. or its affiliates. All rights reserved. Do not distribute. ©Google Inc. or its affiliates. All rights reserved. Do not distribute. 34
Page 16 Page 17
RESTful APIs APIs Explorer
● Programmatic access to products and services ● The APIs Explorer is an interactive tool that lets you easily
○ Typically use JSON as an interchange format try Google APIs using a browser
○ Use OAuth 2.0 for authentication and authorization
● With the APIs Explorer, you can:
● Enabled through the Google Cloud Platform Console
○ Browse quickly through available APIs and versions.
● Most APIs include daily quotas and rates (limits) that can be raised by
○ See methods available for each API and what parameters
request
they support along with inline documentation.
○ Important to plan ahead to manage your required capacity ○ Execute requests for any method and see responses in real
● Experiment with APIs Explorer time.
○ Make authenticated and authorized API calls with ease.
Cloud Console Mobile App Client Libraries

● Manage virtual machines and ● Google Cloud Client Libraries
database instances
○ Community-owned, hand-crafted client libraries
● Manage apps in Google App Engine
● Manage your billing ● Google APIs Client Libraries
● Visualize your projects with a ○ Open source, generated
customizable dashboard ○ Support various languages
■ Java, Python, JavaScript, PHP, .NET, Go, Node.js, Ruby,
Objective-C, Dart
Page 18 Page 19
Agenda Quiz: Service Accounts
Service accounts are used to provide which of the following?

1 Google Cloud Platform Projects
❏ Authentication between Google Cloud Platform services
2 Identity and Access Management (IAM) ❏ Key generation and rotation when used with App Engine and Compute
Engine
3 Interacting with Google Cloud Platform ❏ A way to restrict the actions a resource (such as a VM) can perform
❏ A way to allow users to act with service account permissions
❏ All of the above
4 Quiz
Quiz: Policies and Projects Resources

● Cloud SDK installation and quick start
True or False: If a Google Cloud IAM policy gives you Owner permissions at
https://cloud.google.com/sdk/#Quick_Start
the project level, your access to a resource in the project may be restricted by
a more restrictive policy on that resource. ● gcloud tool guide
https://cloud.google.com/sdk/gcloud/
False: Policies are a union of the parent and the resource. If a parent policy is less restrictive, it
overrides a more restrictive resource policy. ● IAM
https://cloud.google.com/iam/
True or False: All Google Cloud Platform resources are associated with a
project. ● Configuring permissions on Google Cloud Platform
https://cloud.google.com/docs/permissions-overview
True: All Google Cloud Platform resources are associated with a project.
● Google Cloud Platform security
https://cloud.google.com/security/
Page 20 Page 21
Cloud OnBoard
Machine Operations
Compute Networking Big Data Storage
Learning and Tools
Compute Container App Cloud

Beta

Agenda What is Google App Engine?

1 Overview and Customer Stories
● A platform (platform as a service) for
2 Google App Engine Standard Environment
building scalable web applications and
3 Google App Engine Flexible Environment mobile backends
● App Engine makes deployment,
4 Google Cloud Endpoints
maintenance, and scalability easy so
5 Google Cloud Datastore you can focus on innovation
6 Quiz
Page 22 Page 23
IaaS and PaaS Agenda
Compute Engine App Engine 2 Google App Engine Standard Environment

Towards Towards
managed managed
infrastructure services 3 Google App Engine Flexible Environment
IaaS PaaS
Raw compute, storage and Preset run-times
network Java, Go, PHP, Python... 4 Google Cloud Endpoints
More granular control Focus is application logic
5 Google Cloud Datastore

Pay for what you use
Pay for what you allocate Less management overhead
More management overhead 6 Quiz
App Engine Standard Environment

“App Engine enabled us to focus on developing the
Snapchat
application. We wouldn’t have gotten here without the
ease of development that App Engine gave us.” ● Easily deploy your applications
Bobby Murphy, CTO ● Autoscale workloads to meet
demand
Snapchat sends
● Economical
700 million
○ Free daily quota
○ Usage based pricing
photos and videos each day Google App Engine Small team is able ● SDKs for development, testing and
scaled seamlessly to innovate quickly
during growth to and expand deployment
millions of users globally
Page 24 Page 25
App Engine Standard Environment: requirements Agenda
● Specific versions
of Java, Python, PHP, and Go are 2 Google App Engine Standard Environment
supported
3 Google App Engine Flexible Environment
● Need to conform to sandbox constraints: 4 Google Cloud Endpoints
○ No writing to local file system
○ Request timeouts at 60 seconds 5 Google Cloud Datastore
○ Limit on 3rd-party software installations
6 Quiz
Example App Engine Standard Workflow - App Engine Flexible Environment

Web Applications
App Engine automatically App Engine can access a
● Build, deploy containerized apps with a
1 Develop & test the web
3 scales & reliably serves your
web application
variety of services using
dedicated APIs
click
application locally
● Standard runtimes - Python, Java, Go,
Project Memcache
Node.js
App Engine
App Servers
Task
queues ○ No sandbox constraints
Use the SDK to deploy to Application
instances Scheduled
○ Can access App Engine resources
2 App Engine
Application
tasks
● Custom runtime support
instances
Application
Search
○ Any language that supports HTTP
instances Logs requests
Page 26 Page 27
App Engine Standard vs Flexible Environment Google Cloud Endpoints:
distributed API management
Standard Environment Flexible Environment
Instance startup Milliseconds Minutes
● An API console to help you create and
SSH access No Yes (not default)
maintain APIs
● Expose your API using a RESTful interface
Scaling Manual, basic, automatic Manual, automatic
Write to local disk No Yes (ephemeral) ● Control access and validate calls with
Support for 3rd party binaries No Yes JSON Web Tokens and Google API keys
Network access Via App Engine services Yes ○ Identify web, mobile users with Auth0 and
Customizable stack No Yes
Firebase Authentication
Pricing model
After free daily use, pay per instance Pay for resource allocation per hour; ● Generate client libraries
class, with automatic shutdown no automatic shutdown
Agenda Google Cloud Endpoints: supported platforms

1 Overview and Customer Stories ● Supports App Engine Standard or
Flexible Environment, Compute Engine,
2 Google App Engine Standard Environment
Container Engine
3 Google App Engine Flexible Environment ● Use Java or Python open source
Frameworks or any other framework
4 Google Cloud Endpoints
and language
5 Google Cloud Datastore ● Supports iOS, Android, and JavaScript
clients
6 Quiz
Page 28 Page 29
Agenda Google Cloud Datastore: scalable NoSQL DB
● Database designed for application
2 Google App Engine Standard Environment backends
● NoSQL store with automatic scaling
to billions of rows
4 Google Cloud Endpoints ● Fully managed
● Built-in redundancy
5 Google Cloud Datastore ● Supports ACID transactions
6 Quiz
Google Cloud Platform Google Cloud Datastore: benefits
● Schemaless access
Machine Operations
○ No need to think about underlying data
Learning and Tools
structure
● Local development tools
● Includes a free daily quota
Cloud
Storage
Cloud
SQL
Cloud
Datastore
Cloud
Bigtable
● Access from anywhere through a
RESTful interface
Page 30 Page 31
Agenda Resources
1 Overview and Customer Stories ● Overview: App Engine
https://cloud.google.com/appengine/
2 Google App Engine Standard Environment ● DevBytes - Your app, at scale with Google App Engine
https://www.youtube.com/watch?v=ytT2-kL9v2o
● Datastore Concepts Overview
4 Google Cloud Endpoints https://cloud.google.com/datastore/docs/concepts/overview
● Getting started with Google Cloud Datastore API
5 Google Cloud Datastore https://cloud.google.com/datastore/docs/datastore-api-tutorial
6 Quiz
Cloud OnBoard
Quiz
Name 3 advantages of using the App Engine Flexible
Environment over App Engine Standard.
The Flexible Environment allows SSH access, allows disk writes,
and supports third-party binaries. It allows stack customization
and background processes.
True or False: Google Cloud Datastore supports ACID

transactions.
True. Google Cloud Datastore supports ACID transactions.

Page 32 Page 33
Agenda Google Cloud Storage
1 Google Cloud Storage ● High performance, internet-scale,
immutable BLOB (binary large object)
2 Google Cloud Bigtable storage
● Simple administration
3 Google Cloud SQL and Google Cloud Spanner ○ Does not require capacity
management
4 Comparing Storage Options ● Data encryption in-flight and at rest
● Four storage classes give customers
5 Quiz flexibility
Google Cloud Platform Choosing among Cloud Storage classes

Class
Multi-regional Regional Nearline Coldline
Intended for Most frequently Accessed frequently Accessed less than Accessed less than
data that is... accessed within a region once a month once a year
Machine Operations
Compute Networking Big Data Storage Availability SLA 99.95% 99.90% 99.00% 99.00%
Learning and Tools
Access APIs Consistent APIs
Access time Millisecond access
Storage price
Price per GB stored per month
Retrieval price
Cloud Cloud Cloud Cloud Cloud Total price per GB transferred
Storage SQL Spanner Datastore Bigtable
Use cases Content storage and In-region data analytics, Long-tail content, Archiving,
delivery transcoding backups disaster recovery
Page 34 Page 35
Cloud Storage Integration
“Thanks to Google Cloud Platform and the
Google Genomics team, the greatest minds in Startup
Import and
science from around the world will be able to
export scripts,
study trillions of data points in one single
database.” tables images and
general
100 200 PER

object
UPLOADED
RAW
UP TO
BigQuery Compute storage
TERABYTES GENOME
GIGABYTES Engine
of data from more than
1,300 WHOLE MSSNG project
OF DATA
could easily surpass a
GENOMES PETABYTE Object
storage, logs, Cloud Storage Import and
to Google Cloud Storage
Datastore export
WHOLE GENOMES FROM backups tables
10,000
PEOPLE
Making MSSNG world’s
largest single repository
of autism-related DNA
sequencing data
App Cloud SQL
Engine
Agenda
Cloud Storage 1 Google Cloud Storage
Features
2 Google Cloud Bigtable
Offline import (third
Regional buckets Object versioning
party)
3 Google Cloud SQL and Google Cloud Spanner
ACLs
Object lifecycle
Online cloud import
(Cloud Storage
Object change 4 Comparing Storage Options
management notification
Transfer Service)
5 Quiz
Page 36 Page 37
Google Cloud Platform Google Cloud Bigtable: protected, proven
● Replicated storage
Compute Networking Big Data
Machine
Learning
Storage
Operations
and Tools ● Data encryption in-flight and at rest
● Role-based ACLs
● Drives major applications such as
Google Analytics and Gmail
Cloud Cloud Cloud Cloud Cloud
Storage SQL Spanner Datastore Bigtable
Google Cloud Bigtable: managed NoSQL

Overview: Challenges:
● Fully managed, NoSQL, wide-column Data to process: Data in the Consolidated Audit Trail (CAT).
A data repository of all equities and options orders, quotes,
How to process the CAT and organize 100 billion market
events into an “order lifecycle” in a 4 hour window
and events
database service for large-workload Store 6 years (~30PB) of data
applications - Terabytes to petabytes Cloud Bigtable to process and run queries

and tolerate volume increases
● Integrated
○ Accessed using HBase API
○ Native compatibility with big data, 6 BILLION 1.7 GIGs 10 BN 1.7 GIGABYTES
PER SECOND
Hadoop ecosystems WRITTEN
PER SECOND
MARKET EVENTS PER HOUR
WRITTEN PER HOUR

6 TBs BURSTS 10 TERABYTES
PER HOUR
PER HOUR
©Google Inc. or its affiliates. All rights reserved. Do not distribute. 76 Google confidential │ Do not distribute
Page 38 Page 39
Bigtable Access Patterns Agenda
Application API 1 Google Cloud Storage
Data can be read from and written to Cloud Bigtable through a data
service layer like: Managed VMs, the HBase REST Server, a Java
Server using the HBase client. Typically this will be to serve data to
applications, dashboards and data services. 2 Google Cloud Bigtable
Streaming
Data can be streamed in (written event by event) through a variety
of popular stream processing frameworks like:
Cloud Dataflow Streaming, Spark Streaming, and Storm.
Cloud Bigtable
Batch Processing 4 Comparing Storage Options
Data can be read from and written to Cloud Bigtable through batch
processes like: Hadoop MapReduce, Dataflow, Spark. Often,
summarized or newly calculated data is written back to Cloud
Bigtable or to a downstream database. 5 Quiz
Cloud Bigtable Integration Google Cloud Platform

Google Cloud Dataflow
Use Cloud Dataflow connector for Bigtable for batch and
streaming operations in pipelines
Machine Operations
Learning and Tools
Google Cloud Dataproc

Use Bigtable HBase client to integrate Hadoop jobs with
Cloud Dataproc
Cloud Bigtable On-premises, cloud-based Hadoop Cloud Cloud Cloud Cloud Cloud
External Storage SQL Spanner Datastore Bigtable
Use Bigtable HBase client to integrate with Hadoop
service
clusters
Page 40 Page 41
Google Cloud SQL: managed RDBMS Google Cloud Platform
● Offers MySQL and PostgreSQLBeta

databases as a service Compute Networking Big Data
Machine
Storage
Operations
Learning and Tools
● Automatic replication
● Managed backups
● Vertical scaling (read and write)
● Horizontal scaling (read) Cloud Cloud Cloud Cloud Cloud
● Google security Storage SQL Spanner Datastore Bigtable
Cloud SQL Integration Google Cloud SpannerBeta

● Cloud Spanner is a
External
service horizontally-scalable and strongly
consistent relational database.
Cloud SQL can be used with Compute Engine instances Cloud SQL can be used with
App Engine using standard can be authorized to external applications and ● Cloud Spanner supports:
drivers. access Cloud SQL clients. ○ Automatic replication
instances using an external
Standard tools can be used to ○ Strong global consistency
You can configure a Cloud SQL IP address.
administer databases. ○ Managed instances with high
instance to follow an App
Cloud SQL instances can
Engine application.
be configured with a External read replicas can be availability
preferred zone. configured. ○ SQL (ANSI 2011 with extensions)
Page 42 Page 43
Agenda Comparing Storage Options: use cases
1 Google Cloud Storage BLOB NoSQL SQL
2 Google Cloud Bigtable

Cloud Cloud Cloud Cloud SQL Cloud
3 Google Cloud SQL and Google Cloud Spanner Storage Datastore Bigtable Spanner
Good for: Good for: Good for: Good for: Good for:
Structured and Getting started, App “Flat” data, Heavy Web frameworks, Large-scale database
4 Comparing Storage Options unstructured binary
or object data
Engine applications read/write, events,
analytical data
existing applications applications (> ~2 TB)
Use cases: Use cases: Use cases: Use cases: Use cases:
Images, large media User profiles, AdTech, Financial and User credentials, Whenever high I/O,
5 Quiz files, backups product catalog IoT data customer orders global consistency is
needed
Comparing Storage Options: Technical details Agenda

Cloud Cloud Cloud SQL Cloud Spanner Bigtable
Datastore Storage (1st and 2nd
Generation)
1 Google Cloud Storage
Storage type NoSQL, Object Relational SQL Relational SQL NoSQL, wide-column
document (BLOB)
store
Overall Terabytes + Petabytes + up to 500 GB Petabytes Petabytes +
capacity 3 Google Cloud SQL
Unit size 1 megabyte 5 TB / Standard MySQL 10,240 MiB / row Recommended: ~10
/ entity object limits MB per cell, ~100 MB
for all values per row 4 Comparing Storage Options
Transactions Yes No Yes Yes No
Complex
queries
No No Yes Yes No 5 Quiz
88
Page 44 Page 45
Cloud OnBoard
Quiz
You are developing an application that transcodes large video files.
Which storage option is the best choice for your application?
Google Cloud Storage
You manufacture devices with sensors and need to stream huge

amounts of data from these devices to a storage option in the cloud.
Which Google Cloud Platform storage option is the best choice for your
application?
Google Cloud Bigtable

Resources Agenda
● Overview: Cloud Storage
https://cloud.google.com/storage/
1 Introduction to Containers
● DevBytes - File storage in the cloud
https://www.youtube.com/watch?v=vyIap827rHs
2 Kubernetes
● Cloud SQL: Features, case studies, pricing, & documentation
https://cloud.google.com/sql/
● Getting started with Google Cloud SQL
3 Google Container Engine
https://cloud.google.com/sql/docs/quickstart
● Overview of Cloud Bigtable 4 Quiz
https://cloud.google.com/bigtable/docs/overview
Page 46 Page 47
What is a Container? Agenda
● Virtualization at the operating

Virtual Machine
system layer 1 Introduction to Containers
● Separates operating system
from application code and
2 Kubernetes
dependencies Container Container Container
Code and libraries Code and libraries Code and libraries 3 Google Container Engine
● Isolates individual processes Shared libraries
● Popular implementations 4 Quiz

Operating system
include Docker and rkt
Why Use Containers? Kubernetes (“k8s”)

● Open source container cluster
● Support consistency across development, testing, and
orchestration system
production environments
○ Automates deployment, scaling, and
● Loose coupling between application and operating system operations for container clusters
layers
● Based on Google’s experience over
● Much simpler to migrate workloads between on-premises 10+ years
and cloud environments
● Built for a multi-cloud world:
● Support agile development and operations
○ Public, private, hybrid
Page 48 Page 49
Kubernetes eases application maintenance Agenda
● Workload portability
○ Run in many environments, across cloud 1 Introduction to Containers
providers
○ Implementation is open and modular 2 Kubernetes
● Rolling updates
○ Upgrade application with zero downtime 3 Google Container Engine
● Persistent storage
4 Quiz
○ Abstracts details of how storage is provided
from how it is consumed
©Google Inc. or its affiliates. All rights reserved. Do not distribute. 99 ©Google Inc. or its affiliates. All rights reserved. Do not distribute.101
Kubernetes makes applications more elastic Google Cloud Platform
● Multi-zone clusters
○ Run a single cluster in multiple zones
Machine Operations
● Load balancing
Learning and Tools
○ External IP address routes traffic to correct

port
● Autoscaling Compute Container App Cloud

○ Automatically adapt to changes in workload Beta
©Google Inc. or its affiliates. All rights reserved. Do not distribute.100 ©Google Inc. or its affiliates. All rights reserved. Do not distribute.102
Page 50 Page 51
IaaS and PaaS Why use Google Container Engine?
● Decouples operational, development

concerns
Compute Engine Container Engine App Engine
Towards Towards
managed managed ● Manages and maintains
infrastructure services
IaaS PaaS ○ Logging, health management, monitoring
Raw compute, storage and Preset run-times
network Java, Go, PHP, Python... ● Easily update Kubernetes versions as
More granular control Focus is application logic
they are released
Pay for what you use
Pay for what you allocate Less management overhead
More management overhead
Google Container Engine manages GKE’s complementary services

and runs containers
● Fully managed cluster management and ● Google Cloud Container Builder

orchestration system for running containers ○ Create Docker container images from
○ Based on Kubernetes app code in Google Cloud Storage
○ Uses Compute Engine instances and resources
● Uses a declarative syntax to manage
applications ● Google Container Registry
○ Declare desired application configuration, ○ Secure, private Docker image storage
Container Engine implements, manage
Page 52 Page 53
“Our platform sometimes has to be deployed on a cluster. How do we enable
Agenda
containers to communicate from different hosts? Google has the answer:
Treeptik
Kubernetes. This awesome tool helps us manage our clusters of containers as if
they were a single system.”
1 Introduction to Containers
2 Kubernetes
scale speed -30% 3 Google Container Engine
4 Quiz
Docker containers automate scalability REST APIs speed provisioning of new Administrative costs reduced by 30%
instances; JAVA applications can be
deployed in minutes
Deploying Apps: Container Engine vs App Engine Quiz

Name two reasons for deploying applications using containers.
Container App Engine App Engine
Consistency across development, testing, production environments; Simpler
Engine Standard Flexible
to migrate workloads; Loose coupling; Agility
Language Any Java, Python, Go Any
True or False: Kubernetes lets you manage container clusters in multiple cloud
support & PHP
providers.
Service model Hybrid PaaS PaaS True.
Primary use case Container-based Web and mobile Web and mobile True or False: GCP provides a secure, high-speed container image storage
workloads applications applications, service for use with Container Engine.
container-based
workloads True.
Page 54 Page 55
Resources Agenda
● Container Engine Overview
1 Google Compute Engine Overview
https://cloud.google.com/container-engine/
● Container Engine tutorials 2 Google Cloud Networking
https://cloud.google.com/container-engine/docs/tutorials
● Kubernetes 3 Operations and Tools
http://kubernetes.io/
4 Comparing Compute Options
5 Quiz
Cloud OnBoard
Machine Operations
Learning and Tools
Compute Container App Cloud

Beta

©Google Inc. or its affiliates. All rights reserved. Do not distribute. ©Google Inc. or its affiliates. All rights reserved. Do not distribute.114
Page 56 Page 57
Google Compute Engine: managed VMs Google Compute Engine: innovative pricing
● Per-minute billing, sustained use
● High CPU, high memory, standard and
discounts
shared-core machine types
● Preemptible instances
● Persistent disks ● High throughput to storage at no extra
○ Standard, SSD, local SSD cost
○ Snapshots ● Custom machine types - Only pay for
● Robust networking features the hardware you need
Google Compute Engine: many VM options Compute Engine reduces render farm
load during periods of peak production
● High CPU, high memory, standard and

Consumes processing Faster rendering time
shared-core machine types power of up to 15,000 means visual designers
Intel cores at peak can get results and make
● Persistent disks rendering times tweaks more quickly
○ Standard, SSD, local SSD saved due to eliminating idle cores Framestore
$300,000+
○ Snapshots during production “quiet times”
● Resize disks, migrate instances with “By adding Compute Engine to our workflow
and allowing our in-house capacity to focus on the studio
no downtime work, everyone’s project gets computing time – and the
creative team can get as imaginative as they want to,
● Instance metadata and startup scripts with fast views of new iterations.”
Page 58 Page 59
Agenda Cloud Virtual Network
1 Google Compute Engine Overview ● Managed networking functionality for
Cloud Platform resources
2 Google Cloud Networking
● Provision Google Cloud Platform
3 Operations and Tools
resources, connect them to each
other and isolate them from one
another in a Virtual Private Cloud
(VPC)
5 Quiz
Google Cloud Platform Cloud Virtual Network’s internetworking features
● Fine-grained networking policies

Machine Operations
● Granular IP address range selection
Learning and Tools
● Routes
● Firewalls
● Virtual Private Network (VPN)
● Cloud Router
Cloud Cloud Cloud Load Cloud
Virtual Interconnect DNS Balancing CDN
Network
Page 60 Page 61
Google Cloud Interconnect Google Cloud Load Balancing: HTTP(s)
● Balance HTTP-based traffic across

Carrier Interconnect Direct Peering CDN Interconnect multiple Compute Engine regions
Enterprise-grade Connect your Allows select CDN providers to ● Global, external IP address routes
connections provided business directly establish direct interconnect links with
by carrier service to Google Google’s edge network at various
traffic
providers locations ● Scalable, requires no pre-warming and
provides resilience, fault tolerance
Google Cloud Interconnect

Google Cloud DNS Google Cloud Load Balancing: TCP/SSL, UDP

● Highly available and scalable DNS
● Spread TCP/SSL and UDP traffic over
○ Translates domain names into IP pool of instances within a Compute
addresses
Engine region
● Create managed zones, then add, edit, ○ Ensures only healthy instances handle
delete DNS records traffic
○ Programmatically manage zones and ● Scalable, requires no pre-warming
records using RESTful API or
command-line interface
Page 62 Page 63
Google Cloud CDN (Content Delivery Network) Google Cloud Platform
● Use Google's globally distributed edge
caches to cache HTTP(S) load
balanced content far closer to your Compute Networking Big Data
Machine
Storage
Operations
Learning and Tools
users than your instances
○ Faster delivery of content to users while
reducing costs
● Cloud CDN uses caches at network Cloud Deployment

Stackdriver
locations to store responses generated
Source Manager
Repositories
by instances
Agenda Google Stackdriver

1 Google Compute Engine Overview ● Integrated monitoring, logging,
diagnostics
2 Google Cloud Networking ● Powerful data, analytics tools
3 Operations and Tools ● Collaborations with PagerDuty, BMC,

Splunk, othersWorks across Google
4 Comparing Compute Options Cloud Platform, Amazon Web Services
● Open source agents, integration
5 Quiz
Page 64 Page 65
Google Stackdriver’s areas of focus Google Cloud Source Repositories Beta
Monitoring Logging
Platform, system, and Platform, system, and ● Fully-featured Git repositories hosted
application metrics application logs
on Google Cloud Platform
Uptime/health checks Log search/view/filter
Dashboards and alerts Log-based metrics ● Supports collaborative development of
Trace Error Reporting cloud apps
Latency reporting and Error notifications
sampling Error dashboard ● Includes integration with Stackdriver
Per-URL latency and debugger
statistics Debugger
Debug applications
Google Cloud Deployment Manager Agenda
1 Google Compute Engine Overview

● Infrastructure management service
● Create a .yaml template describing 2 Google Cloud Networking
your environment and use Deployment
3 Operations and Tools
Manager to create resources
● Provides repeatable deployments 4 Comparing Compute Options
5 Quiz
Page 66 Page 67
Google Cloud FunctionsBeta Agenda
● Create single-purpose functions that 1 Google Compute Engine Overview
respond to events without a server or
runtime 2 Google Cloud Networking
○ Event examples: New instance created, file
added to Cloud Storage 3 Operations and Tools
● Written in Javascript, execute in
managed Node.js environment on
5 Quiz
Comparing compute options Quiz

Service
Compute Container App Engine App Engine Cloud Name 3 robust networking services available to your applications on
Engine Engine Standard Flex FunctionsBeta Google Cloud Platform.
Language Any Any Java, Python, Go Java, Python, JavaScript
support and PHP Node.js, Ruby, PHP,
Cloud Virtual Network, Cloud Interconnect, Cloud DNS, Cloud
.NET core, and Go; Load Balancing, and Cloud CDN.
or supply your own
runtime Name 3 Compute Engine pricing innovations
Service
model
IaaS Hybrid PaaS PaaS Serverless Per-minute billing, custom machine types, preemptible instances.
Use cases General computing Container-based Web and mobile Web and mobile Ephemeral True or False: Google Cloud Load Balancing lets you balance HTTP
workloads workloads applications applications, functions traffic across multiple Compute Engine regions.
container-based responding to
workloads events True.
Page 68 Page 69
Resources Agenda
● Google Compute Engine
https://cloud.google.com/compute/docs/
● Google Cloud CDN 1 Google Cloud Big Data Platform
https://cloud.google.com/cdn/docs/
● Google Cloud Stackdriver 2 Google Cloud Machine Learning Platform
https://cloud.google.com/stackdriver/docs/
● Google Cloud Deployment Manager
https://cloud.google.com/deployment-manager/docs/
3 Quiz
● Google Cloud Source Repositories
https://cloud.google.com/source-repositories/docs/
Cloud OnBoard
Google Cloud Big Data Platform
Reduces integration risk, accelerates time to value
Integrated, NoOps cloud data
platform for building scalable,
secure and reliable data-driven
applications that transform
businesses and user
experiences.
● Faster time-to-value
● Real-time applications
● Access to innovation,
including machine learning
● Completeness
©Google Inc. or its affiliates. All rights reserved. Do not distribute. ©Google Inc. or its affiliates. All rights reserved. Do not distribute.142
Page 70 Page 71
Google Cloud Platform BigQuery: a fully-managed data warehouse
● Provides near real-time interactive

Compute Networking Big Data
Machine
Storage
Operations analysis of massive datasets
Learning and Tools
(hundreds of TBs)
● Query using SQL syntax (SQL 2011)
● Zero administration for performance
and scale
Cloud Cloud Cloud Cloud
BigQuery
Pub/Sub Dataflow Dataproc Datalab
Big Data Services BigQuery runs on Google’s secure,

high-performance infrastructure
● Compute and storage are separated

with a terabit network in between
BigQuery Pub/Sub Dataflow Dataproc
● You only pay for storage and
Analytics database; Scalable & flexible Stream & batch Managed Hadoop
Stream data at 100,000 enterprise messaging processing; Unified and MapReduce, Spark, processing used
rows per second simplified pipelines Pig, and Hive service
● Automatic discount for long term data
storage
Fully Managed, NoOps Services
Page 72 Page 73
BigQuery at UC Denver Personalized medicinal care for Why use Google Cloud Pub/Sub?
each patient
Anschutz Medical
Campus ● Use cases:
○ Building block for data ingestion in Dataflow,
6 million
Genetic makeup, health history and Internet of Things (IoT), Marketing Analytics
treatments of over 6 million patients ○ Foundation for Dataflow streaming
○ Push notifications for cloud-based
applications
○ Connect applications across Google Cloud
Platform (push/pull between Compute Engine
8 hrs 15 min and App Engine)
BigQuery reduced time to execute
complex analyses on the entire set of
patient records to just 15 minutes
©Google Inc. or its affiliates. All rights reserved. Do not distribute.149
Google Cloud Pub/Sub: scalable, reliable messaging Google Cloud Dataflow: managed data pipelines
● Supports many-to-many asynchronous ● Processes data using Compute Engine
messaging instances
○ Application components make ○ Clusters are sized for you
push/pull subscriptions to topics ○ Automated scaling, no instance
● Includes support for offline consumers provisioning required
● Based on proven Google technologies ● Write code once and get batch and
streaming
● Integrates with Cloud Dataflow for data
○ Transform-based programming model
processing pipelines
Page 74 Page 75
Why use Google Cloud Dataflow? Why use Google Cloud Dataproc?
● Use cases: ● Use cases:
○ ETL (extract/transform/load) pipelines to move, ○ Easily migrate on-premises Hadoop jobs to
filter, enrich, shape data the cloud
○ Data analysis - batch computation or continuous
○ Quickly analyze data (like log data) stored in
computation using streaming
Cloud Storage - create a cluster in less than 2
○ Orchestration - create pipelines that coordinate
minutes then delete it immediately
services, including external services
○ Use Spark/Spark SQL to quickly perform data
● Integrates with GCP services like Cloud mining and analysis
Storage, Cloud Pub/Sub, BigQuery, Bigtable ○ Use Spark Machine Learning Libraries (MLlib)
to run classification algorithms
○ Open source Java and Python SDKs
Google Cloud Dataproc: managed Hadoop Google Cloud Datalab:

interactive data exploration
● Fast, easy, managed way to run
● Interactive tool for large-scale data
Hadoop and Spark/Hive/Pig on Google
exploration, transformation, analysis,
Cloud Platform
visualization
● Create clusters in 90 sec or less ● Integrated, open source
● Scale clusters up and down even when ○ Runs on Google App Engine
jobs are running ○ Built on Jupyter (formerly IPython)
Page 76 Page 77
Why use Google Cloud Datalab? Google Cloud Platform
● Create and manage code, documentation,
results, visualizations in intuitive notebook
format Machine Operations
○ Use Google Charts or matplotlib for easy Learning and Tools
visualizations
● Analyze data in BigQuery, Compute Engine, and

Cloud Storage using Python, SQL, and
JavaScript Speech Translate Natural
Cloud Vision API
● Easily deploy models to BigQuery Machine API API Language
Learning API
Agenda Google Cloud Machine Learning Platform

Open source tool to build and run neural network models
● Wide platform support: CPU or GPU; mobile, server, or cloud
● Developed by researchers and engineers at Google Brain
1 Google Cloud Big Data Platform Fully managed machine learning service
● Faster training, better accuracy versus competing systems
2 Google Cloud Machine Learning Platform Cloud ML

● Familiar notebook-based developer experience
● Optimized for Google infrastructure; Integrates with
BigQuery and Cloud Storage
3 Quiz Pre-trained machine learning models built by Google

● Speech: Stream results in real-time, detects 80 languages
● Vision: Identify objects, landmarks, text, content
Machine Learning APIs ● Translate: Language translation including detection
● Natural Language: Structure, meaning of text
Page 78 Page 79
Google Cloud Machine Learning Use Cases Speech API Beta
Structured Data Unstructured Data ● Recognizes over 80 languages and
Classification/ Regression Image Analytics
● Customer churn analysis ● Identify damaged shipments
variants
● Product diagnostics ● Explicit content classification
● Forecasting ● Identify “styles” in images ● Can return text in real-time
Recommendation Text Analytics ● Highly accurate, even in noisy
● Content personalization ● Call center log analysis
● Product X-sells/up-sells ● Language identification environments
● Topic classification
Anomaly Detection
Sentiment analysis
● Access from any device
● Fraud detection
● Asset sensor diagnostics ● Powered by Google’s machine learning
● Log metric anomalies
Vision API Natural Language API

● Analyze images with a simple REST
API ● Uses machine learning models to
reveal structure, meaning of text
○ Face detection, logo detection, label
● Extract information about people,
detection, and so on
places, events mentioned in text
● With the Cloud Vision API, you can: documents, news articles, blog posts
○ Gain insight from images ● Analyze text uploaded in request or
○ Detect inappropriate content integrate with Cloud Storage
○ Analyze sentiment
○ Extract text
Page 80 Page 81
Translate API Agenda
● Translate arbitrary strings between
thousands of language pairs
1 Google Cloud Big Data Platform
● Programmatically detect a document’s
language 2 Google Cloud Machine Learning Platform
● Support for dozens of languages
3 Quiz
Machine Learning APIs Quiz

Enable apps that see, hear, and understand. Name two use cases for Google Cloud Dataproc.
Migrate on-premises Hadoop jobs to the cloud, data
mining/analysis
Name two use cases for Google Cloud Dataflow.
ETL, orchestration
Name three use cases for the Google machine learning platform
Fraud detection, sentiment analysis, content personalization
Page 82 Page 83
Resources Google Cloud Technical Training and Certification
● Google Big Data Platform
https://cloud.google.com/products/#big-data G Suite
G Suite Admin What New with What New with
Fundamentals G Suite Admin G Suite Admin
● Google Machine Learning Platform

https://cloud.google.com/products/#machine-learning Cloud GCP GCP for Architecting with Architecting with
Fundamentals - Systems GCP: Core GCP: Design and
Infrastructure
Core Infrastructure Operations Infrastructure Process
Data & Machine GCP Data

Fundamentals - Engineering
Learning Big Data & ML On GCP
GCP Developing
Application
Fundamentals - Solutions with
Development Core Infrastructure GCP
Cloud OnBoard
Training and Certification Schedule
cloud.google.com/training
©Google Inc. or its affiliates. All rights reserved. Do not distribute. ©Google Inc. or its affiliates. All rights reserved. Do not distribute.
Page 84 Page 85
Google Cloud Self Paced Qwiklabs
1
3 <thankYou!>
7
See you
8
10
next time
11
12
google.qwiklabs.com
13 </thankYou!>
14
15
16
Google Cloud Coursera
coursera.org/googlecloud
Page 86 Page 87
©2017 Google, Inc.

Book Goog Cloudonboard Northam v2

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Book Goog Cloudonboard Northam v2

Uploaded by

Copyright:

Available Formats

<CloudOnBoard>

Welcome to Cloud OnBoard

©Google Inc. or its affiliates. All rights reserved. Do not distribute.

©Google Inc. or its affiliates. All rights reserved. Do not distribute.

Google Cloud Platform enables developers to build, test and

©Google Inc. or its affiliates. All rights reserved. Do not distribute.

• Pioneering data center efficiency

©Google Inc. or its affiliates. All rights reserved. Do not distribute. 9

Google’s network: Innovative, Customer-Friendly Pricing

2 Singapore over 25% of any calendar month

Committed regions for 2017

©Google Inc. or its affiliates. All rights reserved. Do not distribute. 10

Container Cloud Bigtable Cloud Cloud SQL Cloud Cloud

GCP offers a range of compute architectures

Stackdriver helps engineers keep their

Compute Container App Cloud Managed

IaaS Hybrid PaaS Serverless Automated

Toward managed infrastructure Toward dynamic infrastructure

Sub-hour billing, sustained-use discounts, Compute Engine

In addition to innovative pricing, name 3 benefits of using Google

Commitment to environmental responsibility, commitment to

©Google Inc. or its affiliates. All rights reserved. Do not distribute.

● Projects use three identifying attributes:

Projects organize resources Agenda

IAM predefines “primitive roles” IAM Resource Hierarchy

Service Accounts Example: Service Accounts and IAM

granted objectViewer access to

Interacting with Google Cloud Platform Google Cloud SDK

○ gcloud, gsutil (Cloud Storage), bq

● Available as Docker image

Cloud Console Mobile App Client Libraries

Service accounts are used to provide which of the following?

Quiz: Policies and Projects Resources

Compute Container App Cloud

©Google Inc. or its affiliates. All rights reserved. Do not distribute.

Agenda What is Google App Engine?

Compute Engine App Engine 2 Google App Engine Standard Environment

5 Google Cloud Datastore

App Engine Standard Environment

Example App Engine Standard Workflow - App Engine Flexible Environment

Agenda Google Cloud Endpoints: supported platforms

Google Cloud Platform Google Cloud Datastore: benefits

True or False: Google Cloud Datastore supports ACID

©Google Inc. or its affiliates. All rights reserved. Do not distribute.

Google Cloud Platform Choosing among Cloud Storage classes

Access APIs Consistent APIs

Access time Millisecond access

100 200 PER

©Google Inc. or its affiliates. All rights reserved. Do not distribute. 73

3 Google Cloud SQL and Google Cloud Spanner

Google Cloud Bigtable: managed NoSQL

applications - Terabytes to petabytes Cloud Bigtable to process and run queries

WRITTEN PER HOUR

Cloud Bigtable Integration Google Cloud Platform

Google Cloud Dataproc

● Offers MySQL and PostgreSQLBeta

Cloud SQL Integration Google Cloud SpannerBeta

2 Google Cloud Bigtable

Comparing Storage Options: Technical details Agenda

You manufacture devices with sensors and need to stream huge

©Google Inc. or its affiliates. All rights reserved. Do not distribute.

● Virtualization at the operating

● Popular implementations 4 Quiz

Why Use Containers? Kubernetes (“k8s”)