3. Introduction
10gen is the company behind MongoDB –
the leading NoSQL database
Document- General Open-
Oriented Purpose Source
3
4. 10gen Overview
170+ employees 500+ customers
Offices in New York, Palo Alto, Washington
Over $81 million in funding DC, London, Dublin, Barcelona and Sydney
4
6. Global MongoDB Community
41,000+
Monthly Unique Downloads
24,000+
Online Education Registrants
12,000+
MongoDB User Group Members
10,000+
Annual MongoDB Days Attendees
6
8. Relational Database Challenges
Data Types Agile Development
• Unstructured data • Iterative
• Semi-structured • Short development
data cycles
• Polymorphic data • New workloads
Volume of Data New Architectures
• Petabytes of data • Horizontal scaling
• Trillions of records • Commodity
servers
• Millions of queries per
second • Cloud computing
8
16. MongoDB Features
JSON Data Model with Auto-Sharding for
Dynamic Schema Horizontal Scalability
Flexible, Full Index Rich, Document-Based
Support Queries
Built-In Replication and Fast, In-Place
High Availability Updates
Aggregation Framework and GridFS for Large File
Map/Reduce Storage
16
17. MongoDB Monitoring Service (MMS)
Free, cloud-based service for monitoring and alerts
• Charts, custom
dashboards and
automated alerting
• Tracks 100+ metrics –
performance, resource
utilization, availability
and response times
• 10,000+ users
17
18. Best Total Cost of Ownership (TCO)
Developer and Ops Savings
• Less code
• More productive development
• Easier to maintain
Hardware Savings
• Commodity servers
• Internal storage (no SAN)
• Scale out, not up
Software and Support Savings
• No upfront license – pay for value DB Alternative
over time
• Cost visibility for usage growth
18
19. 10gen Products and Services
Subscriptions
Professional Support, Subscriber Edition and Commercial License
Consulting
Expert Resources for All Phases of MongoDB Implementations
Training
Online and In-Person for Developers and Administrators
19
23. Case Study
Stores user and location-based data in MongoDB
for social networking mobile app
Problem Why MongoDB Results
• Relational architecture • Auto-sharding to scale • Focus engineering on
could not scale high-traffic and fast- building mobile app vs.
growing application back-end
• Check-in data growth
hit single-node capacity • Geo-indexing for easy • Scale efficiently with
ceiling querying of location- limited resources
based data
• Significant work to build • Increased developer
custom sharding layer • Simple data model productivity
23
24. Case Study
Relies on a MongoDB-powered, real-time analytics
product for SMBs
Problem Why MongoDB Results
• More than 500,000 • Ability to handle • Shorter feature
websites complex data while development cycles
maintaining high (e.g., 1 week)
• 10 years of complex performance
data • 2.5x faster than MySQL
• Took 1 week for devs to
• Relational database ramp up on MongoDB • High-performance real-
took several days to time analytics to over
process data • Strong community 500,000 SMBs
24
25. Case Study
Stores billions of posts in myriad formats with
MongoDB
Problem Why MongoDB Results
• 1.5M posts per day, • Flexible document- • Initial deployment held
different structures based model over 5B documents and
• Inflexible MySQL, 10TB of data
• Horizontal scalability
lengthy delays for built in • Automated failover
making changes
provides high
• Data piling up in • Easy to use availability
production database • Interface in familiar • Schema changes are
• Poor performance language quick and easy
25
26. Get Involved
Speak at a MongoDB
meetups@10gen.com
User Group
Get MongoDB News
10gen.com/signup
and Updates
26
27. For More Information
Resource Location
MongoDB Downloads www.mongodb.org/download
Free Online Training education.10gen.com
Webinars and Events www.10gen.com/events
White Papers www.10gen.com/white-papers
Case Studies www.10gen.com/customers
Presentations www.10gen.com/presentations
Documentation docs.mongodb.org
Additional Info info@10gen.com
27
29. Case Study
Serves targeted content to users using MongoDB-
powered identity system
Problem Why MongoDB Results
• 20M+ unique visitors • Easy-to-manage • Rapid rollout of new
per month dynamic data model features
enables limitless
• Rigid relational schema growth, interactive • Customized, social
unable to evolve with content conversations
changing data types throughout site
and new features • Support for ad hoc
queries • Tracks user data to
• Slow development increase
cycles • Highly extensible engagement, revenue
29
30. Case Study
Uses MongoDB to safeguard over 6 billion images
served to millions of customers
Problem Why MongoDB Results
• 6B images, 20TB of • JSON-based data • 5x cost reduction
data model
• 9x performance
• Brittle code base on top • Agile, high improvement
of Oracle database – performance, scalable
hard to scale, add • Faster time-to-market
features • Alignment with
Shutterfly’s services- • Dev cycles in weeks vs.
• High SW and HW costs based architecture tens of months
30
31. Case Study
Stores one of world’s largest record repositories
and searchable catalogues in MongoDB
Problem Why MongoDB Results
• One of world’s largest • Fast, easy scalability • Will scale to 100s of TB
record repositories by 2013, PB by 2020
• Full query language
• Move to SOA required • Searchable catalogue
new approach to data • Complex metadata of varied data types
store storage
• Decreased SW and
• RDBMS could not support costs
support centralized data
mgt and federation of
information services
31
32. Case Study
Provides low-latency, high-scale translation
management platform built on MongoDB
Problem Why MongoDB Results
• Old MySQL • Horizontal scale with built- • Simplified scaling and
performance in sharding high-performance
degradation and high architecture
maintenance • High availability with
replica sets • Dramatically improved
• Complex to scale
MySQL developer productivity
• Memory-mapped
• High- architecture for ingesting • Increased uptime
speed, asynchronous content quickly w/out
storage and fast read separate caching layer
requirements
32
33. Case Study
Uses MongoDB to power real-time ad serving
platform
Problem Why MongoDB Results
• Needed costly SQL • Dynamic schema enables • Billions of requests per
architecture to enable continuous algorithm day with sub-ms latency
real-time bidding development and
customer-specific fields • Inexpensive cost per
• Large volumes of data query
and queries • Scalability for massive
data volumes and low • TB of data stored,
• Diverse, evolving latency populated on the fly and
schema queried in real-time
• Visual monitoring with
MMS
33
34. Case Study
Uses MongoDB to underpin social media monitoring
and recommendation engine
Problem Why MongoDB Results
• HBase locked them into • Ease of use and flexibility • Robust queries and
rigid data model, stifling of data model dynamic schema
ability to create enable higher quality
connections between • Powerful indexing and ad recommendations
data sets hoc querying, plus
integrated MapReduce • Bugs fixed in hours
• Single points of failure instead of days
with master/slave • High availability with
topology replica sets • Dramatically improved
uptime
• Up to 1M posts per day
34
35. Case Study
Stores 3.5 TB of data in MongoDB to power
real-time dictionary
Problem Why MongoDB Results
• Performance • Easy to • Migrated 5B records in
roadblocks with MySQL store, locate, retrieve data 1 day, zero downtime
• Massive data ingestion • Eliminated Memcached • Reduced code by 75%
led to database outages while increasing
performance: up to 2M • Sped up document
• Tables locked for tens requests per hour, 8,000 metadata retrieval from
of seconds during words inserted per 30 ms to 0.1 ms
inserts second • Significant cost
• Long runway for scale-out savings, 15% reduction
in servers
35
36. Case Study
Self-service product built on MongoDB enables
real-time analytics for social marketing
Problem Why MongoDB Results
• Need for real-time • Real-time aggregation to • Operational cost
aggregation and adjust campaigns on the savings
analytics fly
• Simplified scale-out to
• Tried SQL, then • Scalability for persistence support 140M
MapReduce – both layer impressions
solutions only handled
periodic data, could not • Ability to store large • Data flexibility to add
scale amounts of data reliably new features and
performance gains
without overhead
36
37. Case Study
Runs unified data store serving hundreds of
diverse web properties on MongoDB
Problem Why MongoDB Results
• Hundreds of diverse • Flexible schema • Developers can focus on
web properties built on end-user features instead
Java-based CMS • Rich querying and support of back-end storage
for secondary index
• Rich documents forced support • Simplified day-to-day
into ill-suited model operations
• Easy to manage
• Adding new data types, replication and scaling • Simple to add new brands,
tables to RDBMS killed content types, etc. to
read performance platform
37
38. Case Study
MongoDB serves as social gaming platform for
hundreds of millions of users
Problem Why MongoDB Results
• Legacy MySQL • Flexible data model • Improved game
hindered development applicable to wide variety performance and end-user
speed, could not scale of use cases experience
• Needed operational • High availability through • Server cost reduction
database that could replica sets on commodity
also handle real-time servers • Accelerated development
analysis and time-to-market
• Size and strength of
• Server sprawl MongoDB community
38
39. Case Study
Delivers agile automated supply chain service to
retailers powered by MongoDB
Problem Why MongoDB Results
• RDBMS poorly- • Document-oriented model • Decreased supplier
equipped to handle less complex, easier to onboard time by 12x
varying data types code
(e.g., SKUs, images) • Grew from 400K records to
• Single data store for 40M in 12 months
• Inefficient use of structured, semi-
storage in RDBMS structured and • Significant cost reductions
(i.e., 90% empty unstructured data on schema design time,
columns) ongoing developer effort,
• Scalability and availability and storage usage
• Complex joins
degraded performance • Analytics with
MapReduce
39
40. Case Study
Runs social marketing suite with real-time
analytics on MongoDB
Problem Why MongoDB Results
• RDBMS could not meet • Ease of use, developer • Decreased app
speed and scale ramp-up development from months
requirements of to weeks
measuring massive • Solution maturity – depth
online activity of functionality, failover • 30M social events per day
• Inability to provide real- stored in MongoDB
• High-performance with
time analytics and write-heavy system • 6x increase in customers
aggregations
supported over one year
• Queuing and logging for
• Unpredictable peak
loads easy search at app layer
40
41. Case Study
Real-time server and website monitoring
solution runs on MongoDB
Problem Why MongoDB Results
• Needed to handle • General purpose DB • MongoDB-first policy
thousands of requests
per second • High-write throughput • 12+ TB ingested per
month
• MySQL resulted in • Scales easily while
millions of rows per maintaining performance • Increased performance,
month, per server decreased disk usage
• Easy-to-use replication
• Difficult to scale MySQL and automated failover • Simplified infrastructure
with replication cuts costs, frees up
• Native PHP and Python resources for dev
drivers
41
42. Case Study
Social e-commerce application built on MongoDB
offers 100M+ products from over 30K brands
Problem Why MongoDB Results
• MySQL could not • Flexible data model to • Boosted developer
accommodate growth handle varying product productivity
attributes
• Significant optimization • Scaled from 5M to
required to tune MySQL • Scalability for global reach 100M products with
performance minimal work
• Ease of maintenance
• Database maintenance • Decreased product
inhibited development • Consistent performance import time by 90%
even when adding data
and new features
42