Building Enterprise Applications on Google's Cloud Platform Chris SchalkDeveloper Advocate @cschalk

Cloud Computing Expo June 8th, 2011

Agenda

● Navigating the various Cloud Technologies● A Brief History of App Engine● App Engine... A Closer Look● App Engine Leaving Preview● Key Partner Success● Google's new Cloud Technologies● Enterprise Cloud Development on Google● Q&A

Navigating the various Cloud Technologies

IaaS

PaaS

SaaS

Source: Gartner AADI Summit Dec 2009

Cloud landscape as Gartner Sees It

Why classifying a PaaS is not so easy

● SaaS, IaaS are easy to understand... But is PaaS?● What is a "Platform"?

○ Is it purely a development platform?○ Does it involve infrastructure? Should it?

Not all "Platforms" are alike

Why classifying a PaaS is not so easy

● What is the end goal of a PaaS?○ To make an admin's life easier?○ Or to make a developer's life easier?

● Easy to build● Easy to maintain● Easy to scale

App Engine's approach to PaaS

Focus on building your app, let us wear the pagers!

A Brief History of App Engine

A Brief History of App Engine

2008 - App Engine Launches● rapid development: python, dev_appserver● automatic scaling● 30s request deadline● APIs: memcache, datastore, urlfetch, mail● free quotas

2009 - Platform Expands● Billing Support● Java Runtime● Task Queue & Cron● XMPP API● Blobstore API● Datastore Replication

A Brief History of App Engine

2010 - Remove Limitations● Datastore Query● AppStats● Mapper API● 10 minute Tasks● Channel API● Always On

2011 - Enterprise● High-Replication Datastore● Prospective Search API● File API● ProtoRPC● ...!

A Brief History of App Engine

Today (1.5.0)● Backends● Pull Queues with REST API● High Replication Datastore as Default

Soon (h2 2011)● Leaving Preview!

App Engine Growth

App Engine LaunchPython

DatastoreMemcachelogs export

2008 2009 2010

JavaDB Import

cron

Batch write/readHttps

Status-Dashboard

Task QueuesXMPP

incoming email

MultitenancyInstance Console

Always Onhi-perf imag

10 min tasks

BlobstoreAppstats

cursorsMapper

2011

Hi-ReplicationDatastore

Channel APIFiles API

Remote APIProsp Search

100,000+ Active Developers per Month

By the Numbers

200,000+Active apps per week

By the Numbers

1.5B+Pageviews per day

By the Numbers

Some App Engine Partners

App Engine... A closer look

Google App Engine

Cloud Development in a Box

● Downloadable SDK● Application runtimes

○ Java, Python● Local development tools

○ Eclipse plugin, AppEngine Launcher

● Specialized application services

● Cloud based dashboard● Ready to scale ● Built in fault tolerance, load

balancing

Specialized Services

BlobstoreImages

Mail XMPP Task Queue

Memcache Datastore URL Fetch

User Service

Language Runtimes

Extended Language support through JVM

● Java● Scala● JRuby (Ruby)● Groovy● Quercus (PHP)● Rhino (JavaScript)● Jython (Python)

Development Tools for App Engine

App Engine Dashboard

Google App Engine Launcher

SDK Console

Google Plugin for Eclipse

Demo

Building and deploying apps with Google App Engine

App Engine Leaving Preview

Production App Engine Features

● Full Google Product - 3 year deprecation support● 99.95% SLA for all paid users● Operational and Developer Support available● New business-friendly Terms of Service

● Billing via invoice● Rounded out feature set including Frontend and Backend

capabilities

● New sustainable pricing model

New pricing model Overview

● Usage types:○ Free - similar to today○ Paid - $9/app/mo (+ usage) including SLA○ Premier - $500/account/mo (+ usage) including operational

support● No more CPU hours

○ Charging for size and number of running Frontends/Backends (incorporating CPU + memory)

○ APIs (including Datastore) charged based on operations● Add-ons

○ For example: SSL for custom domains will be available for monthly fee

New App Engine Pricing Model details (tentative)

Resource Unit Unit Cost Free per dayOn-Demand Instances Instance-hour $0.08 24

Reserved Instances Instance-hour $0.05 -Datastore Storage GB / mo $0.24 1GDatastore Operations Operation $0.01 / 10k 50k

Blobstore Storage GB / mo $0.17 5G

Outgoing Bandwidth GB $0.15 1G

Incoming Bandwidth GB $0.10 1G

Email API Recipients $0.03 / 100 100

XMPP API Stanza $0.01 / 1k 1k

Channel API Channel Opened $0.01 / 100 100

What is happening to App Engine for Business?

● Preview was announced at Google IO 2010○ 99.9% SLA○ Premium Developer Support○ Custom domain SSL○ Simplified Pricing model○ Company specific administration console

● Trusted Tester program started in Q3 2010○ Gave early access to hundreds of developers○ Held internal and external workshops and webinars to

gather feedback

What is happening to App Engine for Business?

● Overall feedback from Trusted Testers○ All customers were interested in GAE4B features: SLA,

Support, Hosted SQL, SSL for custom domains○ But..

■ Certain aspects of GAE4B were not fully realistic■ Not all companies were focused purely on

building internal apps■ Per user, per app pricing was not appropriate for

all companies

Key Partner Success

Key Enterprise Adoption

WebFilings’ collaborative, browser-based software dramatically improves the efficiency and effectiveness of financial reporting teams

WebFilings Overview

Launched product mid-March 2010● Hundreds of customers with

hundreds more signing-up every quarter

● Marquee customers from F100● Fastest growing company in our

market

Started building product in May 2008● GAE-based from day one● Greatly simplified development,

deployment, and maintenance● Cost-effective● Supports agile development

WebFilings Customers

http://www.webfilings.com/customers/valued_customers

Notable customers:● Procter & Gamble● Delta Airlines● Valero Energy● Sprint/Nextel● eBay ● United Airlines● Dish Network● ...

Recently signed:● One of the world's largest

banks● The world's largest retailer● One of the largest US

auto manufacturers

Massive Scalability Success

Official Royal Wedding on App Engine

On Wedding day, served: ● Up to 32k requests per second● 37.7 million pageviews ● 13.7 million visitors

http://goo.gl/F1SGc

Google's new Cloud Technologies

New Google Cloud Technologies

● Google Storage○ Store your data in Google's cloud

● Prediction API○ Google's machine learning tech in an API

● BigQuery○ Hi-speed data analysis on massive scale

● SQL Service○ Relational Database in the cloud

What Is Google Storage?

● Store your data in Google's cloud○ any format, any amount, any time

● You control access to your data○ private, shared, or public

● Access via Google APIs or 3rd party tools/libraries

Google Storage Benefits

High Availability, Performance, Scalability Backed by Google infrastructure

Strong Security and Privacy Control access to your data

Easy to UseGet started fast with Google & 3rd party tools

Google Storage powers Google Services

● Content hosting - images, data, etc.

● Sharing

● Data Import

● Computation

Haiti Relief Imagery USPTO data

Partner Reporting

Google BigQuery

Google Prediction API

Partner Reporting

Prediction API

● Google's sophisticated machine learning algorithms● Available as an on-demand RESTful HTTP web service

Three Steps to a Smart App:

1. Upload

2. Train

Upload your training data toGoogle Storage

Build a model from your data

Make new predictions

3. Predict

"Auto-classification and sentiment analysis of magazine articles and blog

posts" "Predict profitability (+, or -) of insurance accounts compared to historical outcomes."

"Supply chain product predictions based on historical point of sale and invoice data broken down by distributor, product and store #."

"predict retail product recommendations based on our online purchase/browse behavior"

Actual customer use cases

CustomerSentiment

TransactionRisk

Supply Chain

Recommend Products

BigQuery - Interactive Analysis of Big Data

● Scalable: Billions of rows● Fast: Response in seconds● Simple: Queries in SQL and offered as a web service

1. Upload

2. Import

Upload to Google Storage

Import data into BigQuery Table

Execute queries via APIs

3. Query

Three Steps to a Smart App:● Access via RESTful API, Web UI, BigQuery

Shell, and Spreadsheet

● Example: Query Wikimedia revision history (200M rows, 20G of data.) in < 10s

Google SQL Service

● A fully managed hosted SQL solution

● Millions of turn-key small, mid-size SQL Databases● Fully hosted, native MySQL● Multi-homed● Engine agnostic - MySQL...

Demo

Working with Google Storage, Prediction API...

Enterprise Cloud Development on Google

The Google Enterprise Development Platform

Google Apps Centralized Admin

Enterprise Firewall

Enterprise Data Authentication Enterprise Services User Management

App Engine / CloudGoogle Apps for Business

Google Apps Marketplace

Buy from others Buy from Google Build your own

Google Storage Prediction, BigQuery

and more!

Apps You Build1. Google Apps2. Third party Apps:

Google Apps Marketplace3. ________

Google App Engine

IaaS

PaaS

SaaS

Build your own Apps with App Engine

App Engine/Cloud Resources

Google App Engine ● http://code.google.com/appengine

Google Storage● http://code.google.com/apis/storage

Google Prediction● http://code.google.com/apis/predict

Google BigQuery● http://code.google.com/apis/bigquery

Q&A

Chris SchalkDeveloper Advocate @cschalk

Cloud Computing Expo June 8th, 2011