Thomas Rückstieß - Building your first app; an Introduction to MongoDB

This talk will introduce the philosophy and features of the open source, NoSQL database MongoDB. We’ll discuss the benefits of the document-based data model that MongoDB offers by walking through how one can build a simple app to store books. We’ll cover inserting, updating, and querying the database of books. The second part of the talk will then dive a little deeper into advanced topics, like schema design, query optimization and indexing.

Transcript

1. Technical Services Engineer, MongoDB Thomas Rückstieß #MongoDBDays Building your first app; an Introduction to MongoDB
2. What is MongoDB?
3. MongoDB is a ___________ database •  Document •  Open source •  High performance •  Horizontally scalable •  Full featured
4. Document Database •  Not for .PDF & .DOC files •  A document is essentially an associative array with key/value pairs (possibly nested) { username : “thomas.r", num_logins : 39, last_login : ISODate("2013-10-08T16:46:21Z"), permissions : ["read", "write", "list", "admin"], company : { name : "MongoDB, Inc." location : "Sydney, Australia" } }
5. Open Source •  MongoDB is an open source project •  On GitHub •  Licensed under the AGPL •  Started & sponsored by 10gen (now MongoDB, Inc.) •  Commercial licenses available •  Contributions welcome
6. High Performance •  Written in C++ •  Data serialized as BSON (fast parsing) •  Extensive use of memory-mapped files i.e. read-through write-through memory caching. •  Full support for primary & secondary indexes •  Document model = less work
7. Horizontally Scalable Shard N Shard 3 Shard 2 Shard 1 Horizontally Scalable
8. Full Featured •  Ad Hoc queries •  Real time aggregation •  Rich query capabilities •  Strongly consistent •  Geospatial features •  Support for most programming languages •  Flexible schema
9. Database Landscape Depth of Functionality Scalability & Performance Memcached MongoDB RDBMS Scalability & Performance
10. mongodb.org/downloads
11. $ mongo MongoDB shell version: 2.4.4 connecting to: test > db.test.insert( { text: 'Welcome to MongoDB’ } ) > db.test.find().pretty() { "_id" : ObjectId("51c34130fbd5d7261b4cdb55"), "text" : "Welcome to MongoDB" } Mongo Shell
12. Document Database
13. Terminology RDBMS MongoDB Table, View ➜ Collection Row ➜ Document Index ➜ Index Join ➜ Embedded Document Foreign Key ➜ Reference Partition ➜ Shard
14. Let’s Build a Blog
15. First step in any application is Determine your entities
16. Entities in our Blogging System •  Users •  Articles •  Comments •  Tags •  Categories ( ? )
17. In a relational database app We would start by doing schema design
18. Typical (relational) ERD User ·Name ·Email address Category ·Name ·URL Comment ·Comment ·Date ·Author Article ·Name ·Slug ·Publish date ·Text Tag ·Name ·URL
19. Relational schema design •  Large ERD Diagrams •  Complex “create table” statements •  ORMs to map tables to objects •  Tables just to join tables together •  Lots of revisions until we get it right
20. In a MongoDB based app We start building our app and let the schema evolve
21. MongoDB ERD User ·Name ·Email address Article ·Name ·Slug ·Publish date ·Text ·Author Comment[] ·Comment ·Date ·Author Tag[] ·Value Category[] ·Value
22. Working With MongoDB
23. The Shell
24. > var user = { username: ’thomas.r', first_name: ’Thomas', last_name: ’Rückstieß', } Start with an object (or array, hash, dict, etc)
25. > db test > use blog switched to db blog // syntax is “db.<collection>.<command>” > db.users.insert( user ) Insert the Record No db/collection creation necessary
26. // get one document (don’t care which one) > db.users.findOne() { "_id" : ObjectId("50804d0bd94ccab2da652599"), "username" : ”thomas.r", "first_name" : ”Thomas", "last_name" : ”Rückstieß" } Retrieve the Record again
27. _id •  _id is the primary key in MongoDB •  Automatically created as an ObjectId if not provided •  Automatically indexed •  Any unique immutable value could be used
28. ObjectId •  ObjectId is a special 12 byte value •  Guaranteed to be unique across your cluster •  ObjectId("50804d0bd94ccab2da652599") ts mac pid inc
29. > db.articles.insert( { title: ‘Hello World’, body: ‘This is my first blog post’, date: new Date(‘2013-06-20’), username: ‘thomas.r’, tags: [‘adventure’, ‘mongodb’], comments: [ ] } ) Creating a Blog Article
30. > db.articles.find().pretty() { "_id" : ObjectId("51c3bafafbd5d7261b4cdb5a"), "title" : "Hello World", "body" : "This is my first blog post", "date" : ISODate("2013-06-20T00:00:00Z"), "username" : ”thomas.r", "tags" : [ "adventure", "mongodb" ], "comments" : [ ] } Finding the Article
31. > db.articles.find( { _id : ObjectId("51c3bafafbd5d7261b4cdb5a") } ) .pretty() { "_id" : ObjectId("51c3bafafbd5d7261b4cdb5a"), "title" : "Hello World", "body" : "This is my first blog post", "date" : ISODate("2013-06-20T00:00:00Z"), "username" : ”thomas.r", "tags" : [ "adventure", "mongodb" ], "comments" : [ ] } Finding the Article
32. > db.articles.find( { tags : 'adventure’ } ).pretty() { "_id" : ObjectId("51c3bcddfbd5d7261b4cdb5b"), "title" : "Hello World", "body" : "This is my first blog post", "date" : ISODate("2013-06-20T00:00:00Z"), "username" : ”thomas.r", "tags" : [ "adventure", "mongodb" ], "comments" : [ ] } Querying An Array
33. // the syntax is: “ update ( what, how ) “ > db.articles.update( { _id: ObjectId("51c3bcddfbd5d7261b4cdb5b”) }, { "$push" : { "comments" : { "name" : "Steve Noname", "comment" : "Awesome Post" } } ) Using Update to Add a Comment
34. > db.articles.find( { username: “thomas.r” } ).pretty() { "_id" : ObjectId("51c3bcddfbd5d7261b4cdb5b"), "body" : "This is my first blog post", "comments" : [ { "name" : "Steve Noname", "comment" : "Awesome Post" } ], "date" : ISODate("2013-06-20T00:00:00Z"), "tags" : [ "adventure", "mongodb" ], "title" : "Hello World", "username" : ”thomas.r" } Article with Comment Embedded
35. // remove comment with $pull > var last_comment = { "name" : "Steve Noname", "comment" : "Awesome Post” } > db.articles.update( { _id: ObjectId("51c3bcddfbd5d7261b4cdb5b") }, { $pull : { comments: last_comment } } ) // remove article > db.articles.remove( { _id: ObjectId("51c3bcddfbd5d7261b4cdb5b") } ) Remove Comments / Articles
36. MongoDB Drivers
37. MongoDB has native bindings for over 12 languages
40. MongoDB Drivers •  Official Support for 12 languages •  Community drivers for tons more •  Drivers connect to mongo servers •  Drivers translate BSON into native types •  mongo shell is not a driver, but works like one in some ways •  Installed using typical means (npm, pecl, gem, pip)
41. Indexing in MongoDB
42. Indexes are the single biggest tunable performance factor in MongoDB
43. Why do I need indexes? A brief story… 5 min 30 sec
44. Why do I need indexes? 3h
45. Indexes in MongoDB 7 16 1 2 5 6 9 12 18 21 B-Trees 13 {x: 13} Queries, inserts and deletes: O( log n ) time
46. // create index with ensureIndex, client caches the // index build > db.articles.ensureIndex({ username: 1 }) > db.articles.ensureIndex({ date: –1 }) > db.articles.getIndexes() How Do I Create Indexes? [ { "v" : 1, "key" : { "_id" : 1 }, "ns" : "test.articles", "name" : "_id_" }, { "v" : 1, "key" : { "username" : 1 }, "ns" : "test.articles", "name" : "username_1" }, { "v" : 1, "key" : { "date" : -1 }, "ns" : "test.articles", "name" : "date_-1" } ]
47. // Multiple fields (compound key indexes) > db.authors.ensureIndex({ first_name: 1, last_name: 1 }) // Arrays of values (multikey indexes) > var article = { title : "Hello World", body : "This is my first blog post", ... tags: [’adventure', ’mongodb'] } > db.recipes.ensureIndex({ tags: 1 }) What Can Be Indexed?
48. // Sub-documents > var category = { name: ‘Technical Posts', maintainer: { username: ‘joe.example’, email_address: ’joe_example@myblog.com’ } } db.categories.ensureIndex({ ’maintainer.username': 1 }) db.recipes.ensureIndex({ ’maintainer': 1 }) What Can Be Indexed?
49. Index Options •  Uniqueness constraints (unique, dropDups) •  Sparse Indexes •  Geospatial Indexes •  TTL Collections (expireAfterSeconds)
50. Profiling Slow Operations •  db.setProfilingLevel( n , slowms=100ms ) •  n=0 profiler off •  n=1 record operations longer than slowms •  n=2 record all queries •  db.system.profile.find( ) image: http://www.speareducation.com/
51. > db.articles.find( { tags: ‘mongodb’ } ) .explain( ) { "cursor" : "BasicCursor" , "n" : 42, "nscannedObjects” : 13204, "nscanned" : 13204, ... "millis" : 356, ... } The Explain Plan (without Index)
52. > db.articles.find( { tags: ‘mongodb’ } ) .explain( ) { "cursor" : "BtreeCursor tags_1”, ”isMultiKey" : true, "n" : 42, "nscannedObjects": 42 "nscanned" : 42, ”scanAndOrder" : false, "millis" : 0, ... } The Explain Plan (with Index)
53. Learn more about MongoDB
54. Manual: docs.mongodb.org
55. MongoDB Munich, Oct. 14
56. Online Training at MongoDB University
57. thomas@mongodb.com @tomonezero Thomas Rückstieß #MongoDBDays Questions ? MongoDB Munich October 14, Hilton Munich City www.mongodb.com/events/ mongodb-munich-2013