Java Web Programming Using Cloud Platform Training 18-22 Feb 2013 Module 10: Google App Engine and Heroku

1. Module 10:
Java Cloud Platform
Dr.Thanachart Numnonda
Dr.Thanisa Kruawaisayawan
www.imcinstitute.com
18-22 February 2013

2. Objectives

What is Cloud Computing?

What is Google App Engine?

Google App Engine for Java

Google App Engine Development cycle

Heroku Introduction
2

3. What is Cloud Computing?
3

4. Cloud Computing : Definition
(Wikipedia)
Cloud Computing is Internet-based computing,
whereby shared resources, software, and information
are provided to computers and other devices on
demand, like the electricity grid.
4

5. Cloud Computing Characteristics

Massive, abstracted infrastructure

Dynamic allocation, scaling, movement of
applications

Pay per use

No long-term commitments

OS, application architecture independent

No hardware or software to install
5

6. Grid to Cloud Evolution
6

7. Web 2.0 & Cloud Computing

Web 2.0 concentrates on the private user and clouds
are decscendents of data centers which services the
enterprise.

Web 2.0 promotes SaaS.

Web 2.0 needs massive scaling technologies.

User centric Web 2.0 companies (Twitter,
Slideshare) are relying on Cloud Services.
7

8. ISP to Cloud Evolution
8

9. Software as a Service (SaaS)

SaaS is at the highest layer and features a
complete application offered as a service, on-
demand,

via multitenancy — meaning a single instance of
the software runs on the provider’s infrastructure
and serves multiple client organizations.
9

10. Platform as a Service (PaaS)

The middle layer, or PaaS, is the encapsulation of
a development environment abstraction and the
packaging of a payload of services

PaaS offerings can provide for every phase of
software development and testing, or they can be
specialized around a particular area, such as
content management
10

11. Infrastructure as a Service (IaaS)

IaaS is at the lowest layer and is a means of
delivering basic storage and compute capabilities
as standardized services over the network.

Servers, storage systems, switches,routers, and
other systems are pooled (through virtualization
technology, for example) to handle specific types
of workloads — from batch processing to
server/storage augmentation during peak loads.
11

12. 12

13. Deployment Model

Public Cloud: provider refers to the cloud
platform that targets any types of customers.

Private Cloud: infrastructure that’s hosted
internally, targeting specific customers or
sometimes exclusively within an organization.

Hybrid Cloud: the combination of public and
private clouds, or sometimes on-premise services.
13

14. IaaS & PaaS: Developer's Perspectives

IaaS normally provides up to O/S level as your
choice; for example Amazon Web Services
(AWS) offers several types of Operating Systems
such as Windows Server, Linux SUSE, and Linux
Red Hat. Developer need to install own
middleware, database, etc.

PaaS, given that the database server, VM, and
web server VM are readily provisioned,
14

15. Setting Up App in IaaS
Source:http://acloudyplace.com/2012/01/comparing-iaas-and-paas-a-developers-perspective/ 15

16. Setting Up App in PaaS
Source:http://acloudyplace.com/2012/01/comparing-iaas-and-paas-a-developers-perspective/ 16

17. PaaS for Java

Amazon Elastic Beanstalk

CloudBees

Cloud Foundry

Google App Engine

Heroku for Java

Red Hat OpenShift
17

18. PaaS for Java: Comparison
18

19. PaaS for Java: Comparison
19

20. What is Google App Engine?
20

21. Google App Engine : Definition
(Wikipedia)
It is a platform for hosting web applications in Google-
managed data centers. It is cloud computing
technology which virtualizes applications across
multiple servers and data centers.
21

22. Google App Engine

Running your web application in Google
infrastructure

Support different runtime environments

Java (JRE 6 with limitation, Servlet 2.5, JDO,
JPA)

Python (2.5.2)

Apps run in sandbox.

Automatic scaling and load balancing

No server restart, no network issues
22

23. Hosting Java web apps traditionally

Not so popular except enterprise

High rates as compared to PHP hosting

Shared Tomcat instance among users

Restrictions on any time deployments due to
shared server

Dedicated hosts works fine but they are costly
23

24. You end up with all this
24

25. 25

26. Google Datacenters at Dallas, Oregon
26

27. GAE Architecture
27

28. GAE Physical Deployment Diagram
28

29. Architecture : Application Server
29

30. Distributed web hosting platform
30

31. Distributed Datastore
31

32. Distributed memcache
32

33. Specialized services
33

34. Google Apps + your apps
34

35. Google App Engine for Java
35

36. GAE/J

Was released on April 08 with Python support.
Java included on August 09
36

37. App Engine for Java : One Year
Source: What’s Hot in Java for App Engine Google Con 2010 37

38. GAE Java Runtime Environment

Java 6 VM

Servlet 2.5 Container

HTTP Session support (need to enable explicitly)

JDO/JPA for Datastore API

JSR 107 for Memcache API

javax.mail for Mail API

javax.net.URLConnection for URLFetch API
38

39. Java Standards on GAE
39

40. Services by App Engine

Memcache API – high performance in-memory
key-value cache

Datastore – database storage and operations

URLFetch – invoking external URLs

Mail – sending mail from your application

Task Queues – for invoking background processes

Images – for image manipulation

Cron Jobs – scheduled tasks on defined time

User Accounts – using Google accounts for
authentication
40

41. Limitations

Programming Model : Application runs in
sandbox and can not
– Write to file system
– Make arbitrary network connections
– Use multiple threads/processes
– Perform long-lasting processing
– Permissions
– Know about other instances/applications

Quotas (Requests, In/Out bandwidth, CPU time,
API calls)
41

42. GAE Datastore
42

43. GAE Datastore

Storing data and manipulation

Based on Bigtable

Bigtable is proprietary and hidden from the app
developers

Not a relational database (No SQL)

GQL (Google Query Language) to query

Stores data as entities

Distribution, replication, load balancing behind
the scene

Need to use JDO/JPA
43

44. User Service : Google Accounts

Google Accounts are encouraged as the preferred
authentication mechanism for App Engine
– It assumes that all users have a Google Account
– Google authentication for private domains isn’t
available yet

Access to Google account data -> email, id

The Development Server simulates Google
Accounts

Access constraints based on roles
44

45. User API : Example
import com.google.appengine.api.users.*;
import com.google.appengine.api.users.*;
UserService userService == UserServiceFactory.getUserService();
UserService userService UserServiceFactory.getUserService();
User user == userService.getCurrentUser();
User user userService.getCurrentUser();
String navBar;
String navBar;
if (user == null) {{
if (user == null)
navBar == "<p>Welcome! <a href="" ++ userService.createLoginURL("/")
navBar "<p>Welcome! <a href="" userService.createLoginURL("/")
+"">Sign in or register</a> to customize.</p>";
+"">Sign in or register</a> to customize.</p>";
}} else {{
else
navBar == "<p>Welcome, "" ++ user.getEmail() ++ "! You can <a href=""
navBar "<p>Welcome, user.getEmail() "! You can <a href=""
+userService.createLogoutURL("/") +"">sign out</a>.</p>";
+userService.createLogoutURL("/") +"">sign out</a>.</p>";
}}
45

46. URLFetch API

Invoking external URLs from your application
over HTTP and HTTPs
import java.net.*;
import java.net.*;
import java.io.*;
import java.io.*;
URL url == new URL("htp://...");
URL url new URL("htp://...");
InputStream inp == new InputStreamReader(url.openStream());
InputStream inp new InputStreamReader(url.openStream());
BufferedReader reader == new BufferedReader(inp);
BufferedReader reader new BufferedReader(inp);
String line;
String line;
while ((line == reader.readLine()) != null) {{
while ((line reader.readLine()) != null)
//do something
//do something
}}
reader.close();
reader.close();
46

47. Mail API

Send emails on the behalf of app administrator to
the Google account use.

You can not receive emails
import javax.mail.*;
import javax.mail.*;
Session session == Session.getDefaultInstance(new Properties(), null);
Session session Session.getDefaultInstance(new Properties(), null);
InternetAddress admins == new InternetAddress("admins");
InternetAddress admins new InternetAddress("admins");
Message msg == new MimeMessage(session);
Message msg new MimeMessage(session);
msg.setFrom(admins);
msg.setFrom(admins);
msg.addRecipient(Message.RecipientType.TO, admins);
msg.addRecipient(Message.RecipientType.TO, admins);
msg.setSubject("subject");
msg.setSubject("subject");
msg.setText("text");
msg.setText("text");
Transport.send(msg);
Transport.send(msg);
47

48. Memcache Service

Distributed in memory cache, better than
DataStore

Key-value pair mapping

Configurable expiration time but

Unreliable might be vanished at any time

Supported Interfaces :
– JACHE (JSR 107: JCACHE – Java Temporary
Caching API)
– The Low-Level Memcache API
48

49. Memcache API : Example
import static java.util.Collections.emptyMap;
import static java.util.Collections.emptyMap;
import javax.cache.*;
import javax.cache.*;
CacheFactory cacheFactory == CacheManager.getInstance().getCacheFactory();
CacheFactory cacheFactory CacheManager.getInstance().getCacheFactory();
Cache cache == cacheFactory.createCache(emptyMap());
Cache cache cacheFactory.createCache(emptyMap());
cache.put(key, value);
cache.put(key, value);
cache.get(key);
cache.get(key);
49

50. Task Queues API

Perform background processes by inserting tasks into
queues.

Instructions need to be mention in file queue.xml, in
the WEB-INF/ dir
import com.google.appengine.api.labs.taskqueue.Queue;
import com.google.appengine.api.labs.taskqueue.Queue;
import com.google.appengine.api.labs.taskqueue.QueueFactory;
import com.google.appengine.api.labs.taskqueue.QueueFactory;
import com.google.appengine.api.labs.taskqueue.TaskOptions;
import com.google.appengine.api.labs.taskqueue.TaskOptions;
// ...
// ...
TaskOptions taskOptions ==
TaskOptions taskOptions
TaskOptions.Builder.url("/send_invitation_task")
TaskOptions.Builder.url("/send_invitation_task")
.param("address", "juliet@example.com")
.param("address", "juliet@example.com")
.param("firstname", "Juliet");
.param("firstname", "Juliet");
Queue queue == QueueFactory.getDefaultQueue();
Queue queue QueueFactory.getDefaultQueue();
queue.add(taskOptions);
queue.add(taskOptions); 50

51. Cron Jobs

Up to 20 scheduled tasks per app

Cron jobs (scheduled tasks) supported in cron.xml
in WEB-INF dir

Schedule instructions contain Englis-like format
<?xml version="1.0" encoding="UTF-8"?>
<?xml version="1.0" encoding="UTF-8"?>
<cronentries>
<cronentries>
<cron>
<cron>
<url>/listbooks</url>
<url>/listbooks</url>
<description>Repopulate the cache every day at
<description>Repopulate the cache every day at
5am</description>
5am</description>
<schedule>every day 05:00</schedule>
<schedule>every day 05:00</schedule>
</cron>
</cron>
</cronentries>
</cronentries>
51

52. Images API

Manipulation of images

Transformation of images

Changing image formats
52

53. GAE Development Cycle
53

54. GAE Development Cycle
54

55. Getting Started

The application owner must have a Google
Account to get the tools regardless of language.

Use Java 6 for development.

Eclipse and Netbeans have official plugins.

Both SDKs ship with a Development Web Server
that runs locally and provides a sandbox almost
identical to the real run-time.
55

56. Software Development Kit

App Engine SDK
– Includes web server (Jetty)
– Emulates all the GAE services

SDK includes an upload tool to deploy app to
GAE

Command line tools included.
56

57. Google Plugin for Eclipse
57

58. Development Environment

Development Server

Application lifecycle
management

Eclipse/NetBeans
plugins /
Firefox plugin (GWT)
58

59. Google Plugin for Eclipse
59

60. Development Server
http://localhost:8888
60

61. Development Server Admin Console
http://localhost:8888/_ah/admin
61

62. Deployment Environment

Application is deployed as .war which contains.

Deployment is integrated in IDE

Deploy multiple version of the application at the
same time

Your app lives at
– <app_id>.appspot.com or
– Custom domain with Google Apps
62

63. Running your app on Google
http://<version>.<appid>.appspot.com/some/path
63

64. Managing Applications

Administration Console
http://appengine.google.com/a/yourdomain.com

Application Dashboard

Multiple application versions

Analyzing log files (including admin)

Analyzing resource usage
64

65. GAE Dashboard
65

66. 66

67. Heroku

Platform as a Service for professional
apps developers

Forget servers

Run anything

See everything

Trust & Managed
67

68. Forget Servers

Agile deployment for Ruby, Node.js, Clojure,
Java, Python, and Scala.
68

69. 69

70. 70

71. Resources

Google App Engine at a glance, Stefan Christoph

Developing Java Based Web Applications in

Google App Engine, Tahir Akram, Dec. 2009

Google App Engine, Patrick Chanezon, Mar 2010
71

72. Thank you
thanachart@imcinstitute.com
www.facebook.com/imcinstitute
www.imcinstitute.com
72