Be warned Niklas Gustavsson

Transcription

1 1 Niklas Gustavsson Be warned CouchDB, Slide 2

2 2 Won't replace your relational database You (probably) won't be using it any time soon CouchDB, Slide 3 Forget about relational databases for a while CouchDB, Slide 4

3 3 Open source project started by Damien Katz Soon to be an Apache project CouchDB, Slide 5 How does it work? CouchDB, Slide 6

4 4 Document oriented Documents are JSON {firstname: Niklas, lastname: Gustavsson } Schema less CouchDB, Slide 7 Views Pre-computed, indexed table Incrementally updated Written in JavaScript CouchDB, Slide 8

5 5 // sort and retrieve documents by first name function (doc) { map(doc.firstname, doc); } // retrieve squares by size function (sq) { map(sq.width * sq.height, {color: sq.color}); } CouchDB, Slide So why should I care? CouchDB, Slide 10

6 First reason Sometimes availability trumps consistency CouchDB, Slide 11 CAP theorem - pick two: Consistency Availability Partitioning CouchDB, Slide 12

7 7 Eventual Consistency CouchDB, Slide 13 eventually_consistent.html NODES the number of nodes that store a replica WRITES the number nodes that confirm a commit READS the number of nodes that are contacted at a read operation WRITES+READS > NODES : strong consistency CouchDB, Slide 14 eventually_consistent.html

8 8 Synchronous DB replication NODES = 2 WRITES = 2 READS = > 2! Strong consistency CouchDB, Slide 15 eventually_consistent.html Asynchronous DB replication NODES = 2 WRITES = 1 READS = > 2! Eventual consistency CouchDB, Slide 16 eventually_consistent.html

9 9 What if NODES is 10 or 100 or 1000? Every increase in WRITES means less chance for a write to succeed CouchDB, Slide 17 Read more eventually_consistent.html CouchDB, Slide 18

10 Second reason: Moore has changed tactic CouchDB, Slide 19 CPUs are getting slower but more plenty The Free Lunch is over - CouchDB, Slide 20

11 11 We need to start parallelizing our tasks Concurrency in Java is really, really hard Share nothing Erlang processes and messaging Map/reduce CouchDB, Slide 21 map/reduce Used for computing views in CouchDB CouchDB, Slide 22

12 12 Read more CouchDB, Slide Third reason Web friendly CouchDB, Slide 24

13 13 REST - HTTP used right CouchDB, Slide 25 REST based API Everything is a resource Every resource has a URL Every resource has the same uniform interface Links guides through states Stateless CouchDB, Slide 26

14 14 REST freebies API easily usable from any platform (AJAX, Java, Ruby, COBOL, Powerpoint...)" Works with existing infrastructure Caches, proxies, firewalls Optimistic locking CouchDB, Slide 27 Read more CouchDB, Slide 28

15 15 CouchDB, Slide 29 I'm getting sleepy, show me the goodies! Adding a to do var todotext = $("todoinput").value; this.db.save({text: todotext }); CouchDB, Slide 30

16 16 Adding a tag if(!todo.tags) { todo.tags = [tag]; } else { todo.tags[todo.tags.length] = tag; } this.db.save(todo); CouchDB, Slide 31 The magic of save()" this.save = function(doc, options) { if (doc._id == undefined) { xhr.open("post", this.url); } else { xhr.open("put", this.url + doc._id); } xhr.send(doc.tojsonstring()); } CouchDB, Slide 32

17 17 Questions?? CouchDB, Slide 33 Attributions WHY? - Deep mud / Banana - Forget me not - CouchDB, Slide 34