Mongo Seattle - The Business of MongoDB
Download as KEY, PDF2 likes2,190 views
This document discusses strategies for promoting MongoDB adoption. It recommends understanding MongoDB's features, costs, and use cases before promoting it to potential users. It suggests identifying pain points like complex joins that MongoDB could alleviate for "developers" and performance needs that its MapReduce could address for "managers." Examples of successful MongoDB migrations are presented to convince "skeptics." The goal is to alter perceptions and infect target organizations with MongoDB by solving real problems rather than just touting its features.
![“
We currently have a manual failover
setup with a single master and slave.
The master has 72GB RAM and the
slave is in a different DC. Given disk
space limits, we are in the final stages of
migrating to using automated replica
pairs with manual sharding across 4
database servers [...]
http://blog.boxedice.com/2010/02/28/notes-from-a-production-mongodb-deployment/](https://image.slidesharecdn.com/mongoseattle-100807002635-phpapp02/85/Mongo-Seattle-The-Business-of-MongoDB-27-320.jpg)
![“
Displaying a photo submission form
now requires a single lookup. [...] In a
relational database, embedded
documents would typically be stored
across several tables, requiring
expensive joins. But in a NoSQL store,
embedded documents are surprisingly
efficient and yet let us apply a structure
to the data.
http://open.blogs.nytimes.com/2010/05/25/building-a-better-submission-form/](https://image.slidesharecdn.com/mongoseattle-100807002635-phpapp02/85/Mongo-Seattle-The-Business-of-MongoDB-44-320.jpg)
Ad
More Related Content
What's hot (20)
Viewers also liked (20)
Ad
Similar to Mongo Seattle - The Business of MongoDB (20)
Ad
More from Justin Smestad (7)
Recently uploaded (20)
Mongo Seattle - The Business of MongoDB
- 1. Justin Smestad @jsmestad
- 5. I love Ruby - http://github.com/jsmestad pivotal_tracker.rb sinatra_warden stale_fish
- 7. i love hosting MongoDB mongo machine hosted MongoDB + data management
- 8. mongo machine hosted MongoDB + data management ❖ MongoDB in the cloud ❖ Metered Service ❖ MongoDB Analytics & Statistics ❖ Optional Enhancements
- 9. mongo machine hosted MongoDB + data management ❖ MongoDB in the cloud ❖ Metered Service ❖ MongoDB Analytics & Statistics ❖ Optional Enhancements
- 10. mongo machine hosted MongoDB + data management ❖ MongoDB in the cloud ❖ Metered Service ❖ MongoDB Analytics & Statistics ❖ Optional Enhancements
- 11. mongo machine hosted MongoDB + data management ❖ MongoDB in the cloud ❖ Metered Service ❖ MongoDB Analytics & Statistics ❖ Optional Enhancements
- 12. “ We have used MongoDB on several production and play projects over the last year. We love it. MongoDB is now the default database for new projects at Collective Idea. http://collectiveidea.com/blog/archives/2010/06/15/reflections-on-mongodb/
- 14. Why use MongoDB?
- 15. Why NOT use MongoDB?
- 16. The Business of MongoDB transitioning from amateur to professional Justin Smestad Mongo Seattle - July 27, 2010 mongo machine
- 17. Using MongoDB on every project is simple. Marketing
- 18. The goal is to alter Perception
- 20. Understand MongoDB Before you do anything...
- 21. know the features ❖ Features BSON, GridFS, Map-Reduce ❖ Durability replication scenarios, production environments ❖ Performance mmap, indexing, concurrent requests ❖ Roadmap search, single-server durability auto-sharding, full-text
- 22. know the features ❖ Features BSON, GridFS, Map-Reduce ❖ Durability replication scenarios, production environments ❖ Performance mmap, indexing, concurrent requests ❖ Roadmap search, single-server durability auto-sharding, full-text
- 23. know the features ❖ Features BSON, GridFS, Map-Reduce ❖ Durability replication scenarios, production environments ❖ Performance mmap, indexing, concurrent requests ❖ Roadmap search, single-server durability auto-sharding, full-text
- 24. know the features ❖ Features BSON, GridFS, Map-Reduce ❖ Durability replication scenarios, production environments ❖ Performance mmap, indexing, concurrent requests ❖ Roadmap search, single-server durability auto-sharding, full-text
- 25. oh and one more factor know the costs
- 26. know the costs ❖ Infastructure server costs, migration timeline ❖ Development training, migration ❖ BusinessRequirements Redundancy & Load Specific
- 27. “ We currently have a manual failover setup with a single master and slave. The master has 72GB RAM and the slave is in a different DC. Given disk space limits, we are in the final stages of migrating to using automated replica pairs with manual sharding across 4 database servers [...] http://blog.boxedice.com/2010/02/28/notes-from-a-production-mongodb-deployment/
- 28. know the costs ❖ Infastructure server costs, migration timeline ❖ Development training, migration ❖ BusinessRequirements Redundancy & Load Specific
- 29. know the costs ❖ Infastructure server costs, migration timeline ❖ Development training, migration ❖ BusinessRequirements Redundancy & Load Specific
- 30. Who are you selling to?
- 32. Understand your target developers - make their job easier managers clients
- 33. Understand your target developers managers - why why why? clients
- 34. Understand your target developers managers clients - on time, on budget
- 36. What’s my angle? everyone is a skeptic
- 37. Migrations take time. be patient, pick your battles.
- 38. “ Currently writing checkins, tips, venues (and various things related to venues) to MongoDB. All writes still go to PostgreSQL as well. Slowly migrating various reads. http://docs.google.com/present/view?id=dhkkqm6q_13gm6jq5fv
- 39. let MongoDB be a result of a larger effort Find a cause
- 40. don’t make this mistake. MongoDB is not a feature
- 41. proven causes ❖ Caching caching layer, its amazingly fast as we know. Use MongoDB as a / Performance ❖ Simplifying joins & relationships get rid of those meaningless the domain ❖ Analytics scream Map Reduce! Analytics & Data Collection ❖ Rich UI so much easier. BSON makes life
- 43. Look what I did! Last chance to convince the skeptics
- 44. “ Displaying a photo submission form now requires a single lookup. [...] In a relational database, embedded documents would typically be stored across several tables, requiring expensive joins. But in a NoSQL store, embedded documents are surprisingly efficient and yet let us apply a structure to the data. http://open.blogs.nytimes.com/2010/05/25/building-a-better-submission-form/
- 45. Info Porn lowers barriers so even managers get it.
- 46. Benefits Why was MongoDB better than _________?
- 47. Future Milestones Why switching now was the right choice
- 48. ❖ Understand Features, Targets, Costs ❖ Hook & the Feature Sell Customers ❖ Infect the technology Present and grow
- 49. ❖ Understand Features, Targets, Costs ❖ Hook & the Feature Sell Customers ❖ Infect the technology Present and grow
- 50. ❖ Understand Features, Targets, Costs ❖ Hook & the Feature Sell Customers ❖ Infect the technology Present and grow
- 51. Alter Perception if it didn’t work. iterate.
- 52. Questions? Justin Smestad mongo machine
Editor's Notes
- #3: Big Business - a year with Microsoft then moved on to Orbitz Adv - marketing & rich media focused applications for Oakley & North Face Consulting - working with small dev shops and TechStars companies.
- #4: Big Business - a year with Microsoft then moved on to Orbitz Adv - marketing & rich media focused applications for Oakley & North Face Consulting - working with small dev shops and TechStars companies.
- #5: Big Business - a year with Microsoft then moved on to Orbitz Adv - marketing & rich media focused applications for Oakley & North Face Consulting - working with small dev shops and TechStars companies.
- #7: All of this was bootstrapping Mongo Machine... insert blurb
- #8: All of this was bootstrapping Mongo Machine... insert blurb
- #13: Collective idea has moved to MongoDB for everything. INSPIRING
- #17: The rest of the conf is technically focused.
- #18: When you think about it, using MongoDB on every project like collective idea does is as simple as marketing.
- #22: Performance - what makes mongo fast? single mongod capabilities? Durability - replication options, recommended environments Features - Why? why is BSON important? How does GridFS work? Useful reasons for Map-Reduce. Future Roadmap - whats coming in 1.5, 1.6, 1.7? things like auto sharding, full text search, single server durability
- #23: Performance - what makes mongo fast? single mongod capabilities? Durability - replication options, recommended environments Features - Why? why is BSON important? How does GridFS work? Useful reasons for Map-Reduce. Future Roadmap - whats coming in 1.5, 1.6, 1.7? things like auto sharding, full text search, single server durability
- #24: Performance - what makes mongo fast? single mongod capabilities? Durability - replication options, recommended environments Features - Why? why is BSON important? How does GridFS work? Useful reasons for Map-Reduce. Future Roadmap - whats coming in 1.5, 1.6, 1.7? things like auto sharding, full text search, single server durability
- #25: Performance - what makes mongo fast? single mongod capabilities? Durability - replication options, recommended environments Features - Why? why is BSON important? How does GridFS work? Useful reasons for Map-Reduce. Future Roadmap - whats coming in 1.5, 1.6, 1.7? things like auto sharding, full text search, single server durability
- #26: amazed at how often this is simply ignored.
- #27: infastructure - can your current servers host it? setup costs, migration costs development - training. do you have a DBA on staff that only does SQL? business specific - medical field, odd requirements / regulations can impact cost
- #28: Boxed Ice has 2 server locations, replication, and 72GB of RAM. Production environments require investment like anything else. Knowing all this up front will save you headaches when you people fight you.
- #29: infastructure - can your current servers host it? setup costs, migration costs development - training. do you have a DBA on staff that only does SQL? business specific - medical field, odd requirements / regulations can impact cost
- #30: infastructure - can your current servers host it? setup costs, migration costs development - training. do you have a DBA on staff that only does SQL? business specific - medical field, odd requirements / regulations can impact cost
- #31: Before you market, you have to understand who you have to target. Who are the decision makers?
- #32: The big three. Each of these have different needs. Cater to those needs. Developers - making things easier for themselves Managers - its all about ROI Clients - budget, getting the job done right the first time
- #33: The big three. Each of these have different needs. Cater to those needs. Developers - making things easier for themselves Managers - its all about ROI Clients - budget, getting the job done right the first time
- #34: The big three. Each of these have different needs. Cater to those needs. Developers - making things easier for themselves Managers - its all about ROI Clients - budget, getting the job done right the first time
- #35: The big three. Each of these have different needs. Cater to those needs. Developers - making things easier for themselves Managers - its all about ROI Clients - budget, getting the job done right the first time
- #36: Development This is not application development, this is customer development. Its time to use that understanding to alter perception.
- #37: Marketing technology is about picking your battles. Everyone is a skeptic and looking for holes in your argument. They want you to fail! That is why understanding mongo is to important, but lets not make it that easy... (next slide)
- #38: Migrations do not happen overnight. You have to wait for the right time to stand on that soapbox.
- #39: Take some queues from Foursquare. Harry Heyman from Foursquare gave an entire talk about Mongo migration. Still waiting for the first post about the cold, MongoDB switch... [will it be you?]
- #40: Waiting for the right feature to come along is crucial AND common.
- #41: Time and time again, I see things like this on a legacy project.... MongoDB is not a standalone feature. You cannot sell technology without a purpose.
- #42: Caching - use it for caching, use mongo’s speed for performance. EA - caches their game feeds Domain model - improve the domain model Analytics - Map-Reduce’s sweet spot is analytics. Its also the reason MongoDB came about. Rich UI - Features like BSON make AJAX requests dead simple. Keep those controllers clean!
- #43: You developed the customer & sold mongo to your business/clients.