MongoDB

Mongo (from "humongous") is a high-performance, open source, schema-free document-oriented database. The character "Mongo" from Mel Brook's 1974 movie Blazing Saddles does not help either. As an aside, the actor who played Mongo, Alex Karras, also played the part of the Dad on that lovable '80s sit-com, Webster! Why MongoDB? 10Gen, the creator of MongoDB, most likely felt the need to include the "humongous" part because there has been some controversy in regards to the word "mongo" having a derogatory meaning in multiple language (including English). So to pay homage to Mongo, the humongous gingerbread man, this presentation has literally been themed with his image. Perfect blend of RDBMS and document storage systems Schemaless Dynamic queries Indexes Replication / Failover Auto-sharding (currently in alpha) FAST Great For: Web sites Caching Logging Scalability Storing large objects Overview: Not So Great For: Highly transactional apps Ad-hoc business reporting Problems needing complex SQL to solve The Basics Same as RDBMS Has many collections (tables) Database Collection Same as RDBMS table Group similar documents into smaller query sets (speed) Can be indexed on 1 or more keys Three ways to relate to other collections: embedded document, foreign keys, or db reference Document Stored inside a collection Each document stores both the key and the value (like a hash in Ruby) Storage (BSON) BSON, which stands for "Binary JSON", is the data storage format for "documents" in Mongo. It allows for storage of datatypes that JSON does not allow such as a Date data format and a BinData data format. Querying From: http://railstips.org/2009/6/3/what-if-a-key-value-store-mated-with-a-relational-database-system # finds all Johns db.collection.find({‘first_name’: ‘John’}) # finds all documents with first_name # starting with J using a regex db.collection.find({‘first_name’: /^J/}) # finds first with _id of 1 db.collection.find_first({‘_id’:1}) # finds possible drinkers (age > 21) db.collection.find({‘age’: {‘$gt’: 21}}) # searches in embedded document author for # author that has first name of John db.collection.find({‘author.first_name’:‘John’}) # worse case scenario, or if you need "or" # queries you can drop down to JavaScript db.collection.find({$where:‘this.age >= 6 && this.age <= 18’}) More Querying :fields (like :select in ActiveRecord) :limit, :offset for pagination :sort ascending or descending count and group (uses map/reduce) $in, $nin, $ne, $all, $gt, $gte, $lt, $size, $where Ruby Language Support C++ Java Perl PHP Ruby Python SpiderMonkey Javascript http://www.mongodb.org/display/DOCS/Drivers Drivers / ORMs Mongo Ruby Driver Active Record Adapter MongoRecord MongoMapper http://github.com/mongodb/mongo-ruby-driver/tree/master http://github.com/mongodb/activerecord-mongo-adapter/tree/master http://github.com/mongodb/mongo-activerecord-ruby/tree/master http://github.com/jnunemaker/mongomapper/tree/master Other Cool Stuff Capped collections Data ages out when the collection is full similar to MemCache The data is automatically kept in the insertion order Example: Upserts GridFS Storage for large objects in MongoDB Stores objects as chunks of data Uses 2 collections to store data 'files' contains object metadata 'chunks' contains the binary chunks "find_or_create" in ActiveRecord Example: myColl.update( { name: "Joe" }, { name: "Joe", age: 20 }, { upsert: true } ); db.createCollection("mycoll", {capped: true, size:100000}) MongoMapper Typecasting Callbacks Validations Per object connection Associations Find all, first, last Custom IDs Dynamic finders (e.g. find_by_name) Per object database Create keys on the fly MongoMapper Example (table) (row) http://railstips.org/2009/6/27/mongomapper-the-rad-mongo-wrapper > person = Person.create({ :first_name => :last_name => 'Mulholland', :twitter_screen_name => 'mully', :age => 0, :born_at => Time.mktime(2009, 8, 28, 1, 1), :active => true, :fav_colors => %w(red green blue)}) => #<Person twitter_screen_name: mully, updated_at: Sat Aug 29 05:02:49 UTC 2009, _id: 537a06444a98b67900000012, notes: , first_name: Jim, fav_colors: redgreenblue, born_at: Fri Aug 28 06:01:00 UTC 2009, age: 0, last_name: Mulholland, active: true, created_at: Sat Aug 29 05:02:49 UTC 2009> > person = Person.last > person.fav_colors=> ["red", "green", "blue"] > person["favorite_animal"] = "dog" > person.save > person.favorite_animal=> "dog" > person["family"] = {:wife => "Wendy", :son => "Ben", :twins => "Sam and Nicole"} > person.save > person.family.class=> OrderedHash > person.family["twins"]=> "Sam and Nicole" Creates and Updates Embedded Documents > address = Address.new({ :address => "1 Main", :city => "Houston", :state => "TX", :zip => "77002"}) > person.addresses << address> person.save > person.addresses.first => #<Address city: Houston, zip: 77002, _id: 421477b04a98c3370000007b, address: 1 Main, state: TX> > Person.find(:first, :conditions=>{"addresses.city" => "Houston"}) => #<Person twitter_screen_name: mully, updated_at: Sat Aug 29 05:57:14 UTC 2009, _id: 421477b04a98c2b90000007b, first_name: Jim, fav_colors: redgreenblue, born_at: Fri Aug 28 06:01:00 UTC 2009, age: 18, last_name: Mulholland, active: true, created_at: Sat Aug 29 05:55:05 UTC 2009> > posts = person.posts.build => #<Post updated_at: , _id: , content: , person_id: 421477b04a98c2b90000007b, created_at: > > posts.content = "This is a post" => "This is a post" > posts.save > person.posts.first.content => "This is a post" Validations Foreign Key Relationships > person2 = Person.new({ :first_name => 'Jim', :twitter_screen_name => 'mully', :age => 'twelve', :born_at => Time.mktime(2009, 8, 28, 1, 1), :active => true, :fav_colors => %w(red green blue)}) > person2.save => false > person2.errors => #<Validatable::Errors:0x1a51814 @errors={:age=>["must be a number"], :last_name=>["can't be empty"]}> What is especially cool about this Mongo is that he comes with his own theme song that could be interpretted that Bonnie Tyler (the singer) is asking for a better database. Isn’t there a white knight upon a fiery steed? Late at night I toss and turn and dream of what I need He’s gotta be strong And he’s gotta be fast And he’s gotta be fresh from the fight He’s gotta be sure And it’s gotta be soon And he’s gotta be larger than life But there is a good Mongo... Compared to other DBs "Go, Mongo, go" "More heat, less foam" Thanks Jim Mulholland Twitter: @mully jim@squeejee.com "Let's crash this party, big fella."