1. Introduction to
M. Khurrum Qureshi
Sr. Software Engineer

2. Node’s Goal is to provide an
easy way to build scalable
network programs.

3. Node.js is NOT another
web framework!
But you can create a web framework using NPM modules.

4. Node.js is…
Web Server
TCP Server
Awesome Robot Controller
Command Line Application
Proxy Server
Streaming Server
VoiceMail Server
Music Machine
Anything that has to deal with high I/O

5. Node.js is
Server Side JavaScript!

6. Node.js is
FUN!

7. Why Node.js?
• Non Blocking I/O
• Based on Chrome’s V8 Engines (FAST!)
• 15,000+ Modules
• Active Community (IRC, Mailing Lists, Twitter,
Github)
• Mac, Linux and Windows (all first class citizens)
• One Language for Frontend and Backend
• JavaScript is the Language of the Web

8. Node.js Good Use Cases
• JSON APIs
Building light-weight REST / JSON api's is something where node.js
really shines. Its non-blocking I/O model combined with JavaScript
make it a great choice for wrapping other data sources such as
databases or web services and exposing them via a JSON interface.
• Single Page Apps
If you are planning to write an AJAX heavy single page app (think
gmail), node.js is a great fit as well. The ability to process many
requests / seconds with low response times, as well as sharing things
like validation code between the client and server make it a great
choice for modern web applications that do lots of processing on the
client.

9. Node.js Good Use Cases
• Shelling out to Unix Tools
With node.js still being young, it's tempting to re-invent all kinds of
software for it. However, an even better approach is tapping into the
vast universe of existing command line tools. Node's ability to spawn
thousands of child processes and treating their outputs as a stream
makes it an ideal choice for those seeking to leverage existing
software.
• Streaming Data
Traditional web stacks often treat http requests and responses as
atomic events. However, the truth is that they are streams, and many
cool node.js applications can be built to take advantage of this fact.
One great example is parsing file uploads in real time, as well as
building proxies between different data layers.

10. Node.js Good Use Cases
• Soft Real Time Applications
Another great aspect of node.js is the ease at which you can develop
soft real time systems. By that I mean stuff like twitter, chat software,
sport bets or interfaces to instant messaging networks.

11. Basic HTTP Server
var http = require('http');
var server = http.createServer(function (req, res) {
res.writeHead(200);
res.end('Hello World');
});
server.listen(4000);

12. Some people use the
core http module to
build their web apps,
most use a framework
like Express
or Connect or Flatiron or Tako or Derby or Geddy or Mojito or …

13. Visit
http://expressjs.com/guide.html
for a detailed guide
on using Express

14. What is Non-Blocking I/O?
And why should I care?

15. Blocking I/
270ms = SUM(user, activities, leaderboard)
// Get User – 20ms
$query = 'SELECT * FROM users WHERE id = ?';
$users = query($query, array($id));
print_r($users);
// Get Activities – 100ms
$query = 'SELECT * FROM activities ORDER BY timestamp LIMIT 50';
$activities = query($query);
print_r($activities);
// Get Leader Board – 150ms
$query = 'SELECT count(points) as total, user_id FROM activities LIMIT 50';
$leader_board = query($query);

16. Non-Blocking I/
150ms = MAX(user, activities, leaderboard)
// Get User – 20ms
var query = 'SELECT * FROM users WHERE id = ?';
db.query(query, [userId], function (err, results) {
console.log(results);
});
// Get Activities – 100ms
var query = 'SELECT * FROM activities ORDER BY timestamp LIMIT 50';
db.query(query, function (err, results) {
console.log(results);
});
// Get Leader Board – 150ms
var query = 'SELECT count(points) as total, user_id FROM activities LIMIT 50';
db.query(query, function (err, results) {
console.log(results);
});

17. The most jarring thing
about Server Side JavaScript
is thinking in callbacks

18. The Node Callback Pattern
awesomeFunction(arg, function (err, data) {
if (err) {
// Handle Error
}
// Do something awesome with results.
});
• Error first then success… ALWAYS!
• Because this is the de-facto standard 99.99999% of the time
you will be able to guess how a Node library will work.

19. Callbacks are the Devil’s Work!
Don’t go down this rabbit hole…
One of the biggest mistakes is to get yourself in
to callback hell by nesting callbacks inside of
callbacks inside of more callbacks.
var userQuery = 'SELECT * FROM users WHERE id = ?';
var activityQuery = 'SELECT * FROM activities ORDER BY timestamp LIMIT 50';
var leaderBoardQuery = 'SELECT count(points) as total, user_id FROM activities LIMIT 50';
db.query(userQuery, [id], function (userErr, userResults) {
db.query(activityQuery, function (activityErr, activityResults) {
db.query(leaderBoardQuery, function (leaderBoardErr, leaderBoardResults) {
// Do something here
});
});
});

20. Avoiding Callback Hell
• Keep your code shallow
• Break up your code into small chunks
• Use a sequential library like async
• Visit http://callbackhell.com

21. Async to the rescue!
var async = require('async');
var db = require(’db');
function getUser (callback) {
var query = 'SELECT * FROM users WHERE id = ?';
db.query(query, [userId], callback);
}
function getActivities (callback) {
var query = 'SELECT * FROM activities ORDER BY timestamp LIMIT 50';
db.query(query, callback);
}
function getLeaderBoard (callback) {
var query = 'SELECT count(points) as total, user_id FROM activities LIMIT 50';
db.query(query, callback);
}
var tasks = [getUser, getActivities, getLeaderBoard];
async.parallel(tasks, function (err, results) {
var user = results[0];
var activities = results[1];
var leaderBoard = results[2];
});

22. Visit
https://github.com/caolan/async
for a detailed guide on using the async module.
Async provides several useful
patterns for asynchronous control
flow including: parallel, series,
waterfall, auto and queue.

23. The Node Package Manager
otherwise know as… NPM
It’s how you harness the
awesomeness of the
Node.js community!

24. Using NPM
It’s standard practice to install modules locally for your current project.
Modules are installed in the ./node_modules in the current directory.
To Install a new module
npm install <module>
To find a module in the NPM repository
npm search <search string>
To list the modules (and their dependencies) in the current project
npm list
To see module details
npm info <module>

25. DON’T INSTALL
MODULES GLOBALLY!
Unless they are tools like node-dev, jake, express, minify-js
OR linked development modules but more on that later.

26. NPM is awesome sauce!
Visit
https://npmjs.org
for more details about NPM and to
browse the current NPM Repository

27. Node.js Modules
• async
• connect
• express
• mongodb-
native-driver
• request
• apn
• ql.io-engine
• pem
• winston
• winston-
mongodb

28. Node.js Modules
• node-sql
• nodemailer
• connect-http-
signature
• http-signature
• underscore
• file-utils
• validator
• mongoskin
• passport.Js
• yql

29. Node.js Modules
• node-gcm
• forever
• mongodb_s3_b
ackup
• nconf
• node-sqlserver
• Socket-io
• generic-pool
• And many
others

30. Deployment Platforms
• Amazon EC2
• Windows Azure
• Heroku
• Joynet
• Nodejistu

31. Questions?

32. Introduction to

33. NoSql
• Non-Relational
• Horizontally Scalable
• Distributed
• Schema-Free
• Open-Source
• Replication Support
• Simple API

34. NoSql Flavours
• Key-value Store.
• Graph
• Big Table
• Document Store

35. mongoDB Overview
• Document Database
– Documents (objects) map nicely to programming language data types.
– Embedded documents and arrays reduce need for joins.
– Dynamic schema
• High Performance
– Embedding makes reads and writes fast.
– Indexes can include keys from embedded documents and arrays.
– Optional streaming writes (no acknowledgments).
• High Availability
– Replicated servers with automatic master failover.

36. mongoDB Overview
• Easy Scalability
– Automatic sharding distributes collection data across machines.

37. mongoDB Data Model
• MongoDB instance hosts a number of databases.
• A database holds a set of collections.
• A collection holds a set of documents.
• A document is a set of key-value pairs.
• Documents have dynamic schema.

38. Document Structure
• Data is stored in the form of JSON data.(Which
internally stored as BSON)
{
"_id" : ObjectId("4ccbfc75bd163019417c27f8"),
"title": “Hello World! ",
"author": {
"firstname": "Joe",
"lastname": "Bloggs"
},
"tags": ["test", "foo", "bar"]
}

39. Key mongoDB Features
• Flexibility
• Power
• Speed/Scaling
• Ease of use

40. Questions?