3.0 Features
Document-Level Concurrency via WiredTiger
Throughput, latency improvements, concurrency
YCSM – improved performance with WiredTiger
Pebble is using WiredTiger is powering their API. WT is fast, 500k users on 1 shard.
WiredTiger will be default in MongoDB 3.2
Storage Engine API in MongoDB 3.0 used by WiredTiger
Saved them 3 years.
Compression in WiredTiger with Snappy
Soundwave (phone social app) – use compression
facebook built RocksDB used by 20% of Parse data sets
One more storage engine
/dev/null engine works with sharding
used for testing
Automation – in cloud via MMS, Ops Manager for on premise
q1media – use MMS to backup production and then use it to populate dev dbs
MMS – monitoring, backup, automation
MongoDB 3.2
1. Rules and Tools
Types, indexes, schema is done by application (dynamic, flexible, polymorphic schema)
Partner feeds bugs
Data exploration
Partial Indexes
filter: {flags: “good”}
Document validation
Mongo Scout
cursor.next()
Dan P. – VP of Engineering
Scout shows visualization, document insertions, charts, list of fields in the collections
db.runCommand( {colMod: “collectionName”, validator: {age: {$gte: 21}}});
db.collectionName.insert({…}) now requires age field with a range
Can define complex validation statements
Coming in Mongodb 3.2. Flexible schema on demand.
Pipeliness – Unix – Mongodb 3.2 Querying
Reporting, push down to the database
Adding more aggregation operators
New Operator: $lookup
show collections
$unwind: flatten documents into 1 output for each array element
$lookup: seems like left outer join
$unwind, $lookup, $group — aggregation
db.collectionName.aggregate(a3)
Pictures
Visualizer
BI Connector will work with any SQL compliant BI Tool
Beta partner with tableau
Tableau remoting calculations to MongoDB
Live reports, charts and analytics
rapid app development of mongodb
The post MongoDB World Day 2, 2015 appeared first on MongoDB Developer.