1. MongoDB and PHP

2. My name is
Steve Francia
@spf13

3. • 15+ years building the
internet
• Father, husband,
skateboarder
• Chief Solutions Architect @
10gen
• Author of upcoming O’Reilly
publication “PHP and
MongoDB”

4. Introduction to
MongoDB

5. Why MongoDB?

6. MongoDB Goals
• OpenSource
• Designed for today
• Today’s hardware / environments
• Today’s challenges
• Easy development
• Reliable
• Scalable

7. • Company behind MongoDB
• AGPL license, own copyrights, engineering
team
• support, consulting, commercial license
revenue
• Management
• Google/DoubleClick, Oracle, Apple, NetApp
• Funding: Sequoia, Union Square, Flybridge
• Offices in NYC, Redwood Shores & London
• 60+ employees

8. A bit of history

9. 1974
The relational database is created

12. 1979

13. 1979 1982-1996

14. 1979 1982-1996 1995

15. Computers in 1995
•Pentium 100 mhz
•10base T
•16 MB ram
•200 MB HD

16. Cell Phones in 2011
•Dual core 1.5 Ghz
•WiFi 802.11n (300+ Mbps)
•1 GB ram
•64GB Solid State

17. How about a DB
designed for today?

18. It started with
DoubleClick

19. Signs something
needed
•doubleclick - 400,000 ads/second
•caching is de rigueur
•complex ORM frameworks
•computer architecture increasing in
complexity
•cloud computing

20. Requirements
We need enough functionality to handle
broad set of use cases
•strong consistency / atomicity
•secondary indexes
•ad hoc queries

21. Trim unneeded
features
Leave out a few things so we can scale
•no choice but to leave out relational
•distributed transactions are hard to
scale

22. Needed a scalable
data model
Considered....
•key/value
•columnar / tabular
•document oriented (JSON inspired)

23. MongoDB philosphy
• Keep functionality when we can (key/
value stores are great, but we need more)
• Non-relational (no joins) makes scaling
horizontally practical
• Document data models are good
• Database technology should run
anywhere VMs, cloud or straight metal

24. MongoDB is:
Application Document
Oriented
{ author: “steve”,
High date: new Date(),
text: “About MongoDB...”,
Performance tags: [“tech”, “database”]}
Fully
Consistent
Horizontally Scalable

25. MongoDB
•JSON Documents
•Querying/Indexing/Updating similar to
relational databases
•Traditional Consistency
•Auto-Sharding

26. Under the hood
•Written in C++
•Available on most platforms
•Data serialized to BSON
•Extensive use of memory-mapped files

27. Database Landscape

28. This has led
some to say
“
MongoDB has the best
features of key/ values
stores, document databases
and relational databases in
one.
John Nunemaker

29. Use Cases

30. CMS / Blog
Needs:
• Business needed modern data store for rapid development and
scale
Solution:
• Use PHP & MongoDB
Results:
• Real time statistics
• All data, images, etc stored together, easy access, easy
deployment, easy high availability
• No need for complex migrations
• Enabled very rapid development and growth

31. Photo Meta-Data
Problem:
• Business needed more flexibility than Oracle could deliver
Solution:
• Use MongoDB instead of Oracle
Results:
• Developed application in one sprint cycle
• 500% cost reduction compared to Oracle
• 900% performance improvement compared to Oracle

32. Customer Analytics
Problem:
• Deal with massive data volume across all customer sites
Solution:
• Use MongoDB to replace Google Analytics / Omniture options
Results:
• Less than one week to build prototype and prove business case
• Rapid deployment of new features

33. Online Dictionary
Problem:
• MySQL could not scale to handle their 5B+ documents
Solution:
• Switched from MySQL to MongoDB
Results:
• Massive simplification of code base
• Eliminated need for external caching system
• 20x performance improvement over MySQL

34. E-commerce
Problem:
• Multi-vertical E-commerce impossible to model (efficiently) in
RDBMS
Solution:
• Switched from MySQL to MongoDB
Results:
• Massive simplification of code base
• Rapidly build, halving time to market (and cost)
• Eliminated need for external caching system
• 50x+ improvement over MySQL

35. Tons more
Pretty much if you can use a RDMBS or Key/Value
MongoDB is a great fit

36. In Good Company

37. MongoDB in PHP

38. Relational made normalized
data look like this

39. Document databases make
normalized data look like this

40. Terminology
RDBMS Mongo
Table, View ➜ Collection
Row ➜ JSON Document
Index ➜ Index
Join ➜ Embedded Document
Partition ➜ Shard
Partition Key ➜ Shard Key

41. Documents to Objects
Documents to Arrays
Tables to Documents
{
PostObject Object(
[title] => MongoDB
title: ‘MongoDB’,
[contributors] => Array(
contributors: [
[0] => PersonObject Object(
{ name: ‘Eliot Horowitz’,
[name] => Eliot Horowitz
[email] => eh@10gen.com
email: ‘eh@10gen.com’ },
)
{ name: PersonObjectMerriman’,
[1] => ‘Dwight Object(
email: ‘dm@10gen.com’ }
[name] => Dwight Merriman
[email] => dm@10gen.com
],
)
model: {
)
relational: false,
[model] => ModelObject Object(
[relational] =>
awesome:=>true
[awesome] 1
})
)
}

42. Documents to Objects
Documents to Arrays
Tables to Documents
{ array(
PostObject Object(
title: ‘MongoDB’, "title" =>=> MongoDB
[title] 'MongoDB',
"contributors" =>=> Array(
[contributors] array(
contributors: [ array( => PersonObject Object(
[0]
{ name: ‘Eliot Horowitz’, 'name' => 'Eliot Horowitz',
[name] => Eliot Horowitz
email: ‘eh@10gen.com’ }, 'email' => 'eh@10gen.com'
[email] => eh@10gen.com
), )
{ name: ‘Dwight Merriman’, array( => PersonObject Object(
[1]
email: ‘dm@10gen.com’ } 'name' => 'Dwight Merriman',
[name] => Dwight Merriman
'email' => 'dm@10gen.com'
[email] => dm@10gen.com
],
) )
model: { ),)
relational: false, "model" =>=> ModelObject Object(
[model] array(
'relational' => false,
[relational]
awesome: true 'awesome' => true
[awesome] 1
} ) )
} )

43. Documents to Objects
Documents to Arrays
Tables to Documents
{ array(
PostObject Object(
title: ‘MongoDB’, "title" =>=> MongoDB
[title] 'MongoDB',
"contributors" =>=> Array(
[contributors] array(
contributors: [ array( => PersonObject Object(
[0]
{ name: ‘Eliot Horowitz’, 'name' => 'Eliot Horowitz',
[name] => Eliot Horowitz
email: ‘eh@10gen.com’ }, 'email' => 'eh@10gen.com'
[email] => eh@10gen.com
), )
{ name: ‘Dwight Merriman’, array( => PersonObject Object(
[1]
email: ‘dm@10gen.com’ } 'name' => 'Dwight Merriman',
[name] => Dwight Merriman
'email' => 'dm@10gen.com'
[email] => dm@10gen.com
],
) )
model: { ),)
relational: false, "model" =>=> ModelObject Object(
[model] array(
'relational' => false,
[relational]
awesome: true 'awesome' => true
[awesome] 1
} ) )
} )

44. MongoDB speaks
PHP

45. This has led
some to say
“
The single best interface
(API) of any php
extension.
Matthew Weier O'Phinney

46. PHP Driver
Installing the driver
$ pecl install mongo
Add to php.ini
extension = mongo.so

47. Connecting to the DB
$connection = new Mongo();
$db = $connection->selectDB('blog');
$posts = $db->post;
If the DB or collection doesn’t exist,
mongoDB will create it on connect.

48. Documents
Blog Post Document
$p1 = array("author" => "roger",
"date" => new MongoDate(),
"text" => "about mongoDB...",
"tags" => array('tech', 'databases')
);
$posts->save($p1);

49. Querying
print_r($posts->findOne());
> db.posts.findOne()
Array(
[_id] => MongoId Object(
> { _id : ObjectId("4c4ba5c0672c685e5e8aabf3"),
[$id] => 4e9796764a18173a17000000
author : "roger",
)
date : "Sat Jul 24 2010 19:47:11",
[author] => roger
text : "About MongoDB...",
[date] => MongoDate Object(
tags : [ "tech", "databases" ] }
[sec] => 1318557302
[usec] => 581000
)
[text] => about mongoDB...
[tags] => Array(
Note: _id is unique, but can be anything
[0] => tech
[1] => databases
you’d like
)
)

50. Querying
print_r($posts->findOne());
> db.posts.findOne()
Array(
[_id] => MongoId Object(
> { _id : ObjectId("4c4ba5c0672c685e5e8aabf3"),
[$id] => 4e9796764a18173a17000000
author : "roger",
)
date : "Sat Jul 24 2010 19:47:11",
[author] => roger
text : "About MongoDB...",
[date] => MongoDate Object(
tags : [ "tech", "databases" ] }
[sec] => 1318557302
[usec] => 581000
)
[text] => about mongoDB...
[tags] => Array(
Note: _id is unique, but can be anything
[0] => tech
[1] => databases
you’d like
)
)

51. Inserting Objects
class PostObj {
public $comments = array();
public function __construct($author, $title, $content)
{
$this->author = $author;
$this->title = $title;
$this->content = $content;
}
}
$posts->insert(new PostObj("Steve",
"My first blog post",
"Hello, World!"));

52. MongoId
An autogenerated primary key
$x = array("foo" => 1, "bar" => 2);
$db->baz->save($x);
var_dump($x['_id']);
Result:
object(MongoId)#4 (1) {
["$id"]=> string(24) "4e9cc76a4a1817fd21000000"
}

53. MongoId
An autogenerated primary key
object(MongoId)#4 (1) {
["$id"]=> string(24) "4e9cc76a4a1817fd21000000"
}
4e9cc76a4a1817fd21000000
|------||----||--||----|
ts mac pid inc

54. Update Operations
$set, $unset, $inc, $push, $pushAll,
$pull, $pullAll, $bit
$change = array('$push' =>
array('comments' =>
array(
'author' => 'Fred',
'created' => new MongoDate(),
'comment' => 'Best post ever!.'
)
)
);
$posts->update(array("_id" => $id), $change);

55. Nested Documents
Array(
{ _id : ObjectId("4c4ba5c0672c685e5e8aabf3"),
[_id] => MongoId Object (
[$id] => 4e9796764a18173a17000000
) author : "roger",
[author] => roger
date : "Sat Apr 24 2011 19:47:11",
[comments] => Array(
text : "About MongoDB...",
[0] => Array(
[author] => Fred
tags : [ "tech", "databases" ],
[created] => MongoDate Object(
[sec] => 1318899392
comments : [ [usec] => 176000
)
{
[comment] => Dumb post.
)
) author : "Fred",
[date] => MongoDate Object(
date : "Sat Apr 25 2010 20:51:03 GMT-0700",
[sec] => 1318557302
[usec] => 581000
) text : "Best Post Ever!"
[tags] => Array(
}
[0] => tech
)
]
[1] => databases
}
)
[text] => about mongoDB...

56. Nested Documents
Array(
{ _id : ObjectId("4c4ba5c0672c685e5e8aabf3"),
[_id] => MongoId Object (
[$id] => 4e9796764a18173a17000000
) author : "roger",
[author] => roger
date : "Sat Apr 24 2011 19:47:11",
[comments] => Array(
text : "About MongoDB...",
[0] => Array(
[author] => Fred
tags : [ "tech", "databases" ],
[created] => MongoDate Object(
[sec] => 1318899392
comments : [ [usec] => 176000
)
{
[comment] => Dumb post.
)
) author : "Fred",
[date] => MongoDate Object(
date : "Sat Apr 25 2010 20:51:03 GMT-0700",
[sec] => 1318557302
[usec] => 581000
) text : "Best Post Ever!"
[tags] => Array(
}
[0] => tech
)
]
[1] => databases
}
)
[text] => about mongoDB...

57. Querying
$posts = $blog->find(array(
"author" => "Roger"));
$commentsByFred = $blog->find(array(
"comments.author" => "Fred"));
$commentsByFred = $blog->find(array(
"comments.author" =>
new MongoRegex("/fred/i")));

58. It’s all about the $
$ instead of >, <, =, etc.
$gt, $gte, $lt, $lte, $eq, $neq, $exists,
$set, $mod, $where, $in, $nin, $inc
$push, $pull, $pop, $pushAll, $popAll
$c->find(array("x" => array('$gt' => 4)));
$c->find(array("x" => array(“$gt” => 4)));

59. It’s all about the $
$ instead of >, <, =, etc.
$gt, $gte, $lt, $lte, $eq, $neq, $exists,
$set, $mod, $where, $in, $nin, $inc
$push, $pull, $pop, $pushAll, $popAll
$c->find(array("x" => array('$gt' => 4)));
$c->find(array("x" => array(“$gt” => 4)));

60. Adjust the behavior
in php.ini or using ini_set()
php.ini
---------------
mongo.cmd = ":"
$c->find(array("x" => array(':gt' => 4)));
or
ini_set("mongo.cmd", ".");
$c->find(array("x" => array('.gt' => 4)));

61. Indexing
including secondary and compound
indexes
$people->ensureIndex(array("age" => 1));
$people->ensureIndex(array(
"name" => -1,
"ts" => -1,
"comments.author" => 1
));

62. More Indexing
// Index nested documents
$posts->ensureIndex( array("comments.author" => 1) )
$posts->find( array("comments.author" => "Fred") )
// Index on tags (multi-key index)
$posts->ensureIndex( array("tags" => 1 ) )
$posts->find( array("tags" => "tech") )
// geospatial index
$posts->ensureIndex( array("author.location" => "2d") )
$posts->find( array("author.location" =>
array( '$near' => array(22,42)));

63. Cursors
$cursor = $c->find(array("foo" => "bar"))
foreach ($cursor as $id => $value) {
echo "$id: ";
var_dump( $value );
}
$a = iterator_to_array($cursor);

64. Paging
$page_num = 3;
$results_per_page = 10;
$cursor = $results->find()
->sort(array("ts" => -1))
->skip($page_num * $results_per_page)
->limit($results_per_page);

65. Rich Documents
{ _id : ObjectId("4c4ba5c0672c685e5e8aabf3"),
line_items : [ { sku: ‘tt-123’,
name: ‘Coltrane: Impressions’ },
{ sku: ‘tt-457’,
name: ‘Davis: Kind of Blue’ } ],
address : { name: ‘Banker’,
street: ‘111 Main’,
zip: 10010 },
payment: { cc: 4567,
exp: Date(2011, 7, 7) },
subtotal: 2355
}

66. Grid FS

67. Storing Files
Under 16mb

68. Storing Big Files
>16mb stored in 16mb chunks

69. Storing Big Files
Works with replicated
and sharded systems

70. A better network FS
• GridFS files are seamlessly sharded & replicated.
• No OS constraints...
• No file size limits
• No naming constraints
• No folder limits
• Standard across different OSs
• MongoDB automatically generate the MD5 hash
of the file

71. Storing Files
$grid = $db->getGridFS();
// The file's location in the File System
$path = "/tmp/";
$filename = "movie.mp4";
// Note metadata field & filename field
$storedfile = $grid->storeFile($path . $filename,
array("metadata" => array("filename" => $filename),
"filename" => $filename));
// Return newly stored file's Document ID
echo $storedfile;

72. Storing Data
$grid = $db->getGridFS();
$filename = "zelda.jpg";
$storedfile = $grid->storeBytes("ÿØÿÉ
€ ^@^PJFIF^@^A^A^A,^G^E^E^G,...",
array("metadata" => array("filename" =>
$filename),
"filename" => $filename));

73. Streaming Data
$gridFS = $db->getGridFS();
// Find image to stream
$image = $gridFS->findOne("zelda.jpg");
// Stream image to browser
header('Content-type: image/jpeg');
echo $image->getBytes();

74. PHP Libraries for
MongoDB

75. Libraries & Frameworks
• Doctrine MongoDB ODM • Zend Framework
• Active Mongo • Lithium
• Mongo Queue • Symfony2
• Morph • Fuel
• Mandango • Fat Free Framework

76. Doctrine MongoDB ODM
• Provides full validation
• Seamless integration with Doctrine ORM
• Useful for hybrid solutions
• Follows same persistence model as Doctrine
• Uses a Document Manager & annotations
• Supports embedded and referenced objects
• Can use mongo’s query interface
• Supports in place updates
https://github.com/doctrine/mongodb-odm

77. Mandango
•Very light and fast
•Event support
•Supports embedded and referenced
objects
•uses the mongo query syntax
•Lacking validation, but events can be
utilized to validate
https://github.com/mandango/mandango

78. Active Mongo
•Uses ActiveRecord (active document)
interface
•Supports full validation
•Supports in place updating
•Uses native query interface
•Supports referenced documents
https://github.com/crodas/ActiveMongo

79. Morph
•Follows the Active Record model
•Has own query object (Morph_Query)
•Doesn’t provide validation
•Doesn’t support embedded or referenced
objects.
https://github.com/a-musing-moose/morph

80. Mongo Queue
An extensible job/message queueing system that
uses mongodb as the persistent storage engine.
• Fully Configurable
• Distributed
• Atomic Locking
• Priority Support
• Worker Timeout Support
• Stable
https://github.com/skiz/mongo_queue

81. http://spf13.com
http://github.com/spf13
@spf13
Questions?
download at mongodb.org
PS: We’re hiring!! Contact us at jobs@10gen.com