Last purchase from Amazon:
Creating Tomorrow’s Schools Today from @richardgerver
better = simple = easy = true;
Easy, simple and fast.
Inspired by the famous Wordpress 5 minutes installer.
#humanthings Working on its documentation.
Scaling with MongoDB (not how to)
More links and learning on nosql: comparing Cassandra and MongoDB.
http://stackoverflow.com/questions/2892729/mongodb-vs-cassandra may 2010
- Both databases perform well on reads where the hot data set fits in memory. Writes are more problematic in MongoDB. See Global write lock.
- Scalability: MongoDB for single node as cassandra offers more control on replication. My knowledge and tests on scalability with Cassandra confirmed how easy to scale is, don’t know about MongoDB - jr.
- For analytics: MongoDB provides a map/reduce implementation; Cassandra provides native Hadoop.
If you’re presently using JSON blobs, mongo is an insanely good match for your use case, given it uses BSON to store the data. You’ll be able to have richer and more queryable data than you would in your present database. This would be the most significant win for Mongo.
Read Richard K. answer for more about this - jr.
MongoDB: Boris Filipov made a great work comparing reads using node and php drivers.
Cassandra: A very interesting video about scaling How to set up a 4 node cluster in under 2 minutes
Make no little plans; they have no magic to stir men’s blood.
“Try it out” @ http://www.mongodb.org/, the kind of things I love from UX & UI design. <3
Simple thoughts on running databases on Amazon Web Services and MongoDB
Simple DB (beta): http://aws.amazon.com/en/simpledb/
Intervention 0, the best solution if your need to have a high available no relational database without setup time or other costs. It have several limitations on scalability. As you will find in AWS case studies business use SimpleDB for operational data and RDS for mission critical data (see Flipboard).
$0,140/h Virginia & Oregon and $0,154/h North Califormia. This is $100,8 and $110,88 monthly. No data transfer costs if the the calls are made from other Amazon WS. Storage: GB/month is both $0,250 or $0,275. Pretty pretty low cost.
It’s a solution for operational data as it have imitations for growth forecast scenarios like 10GB per domain and performance limitations. See Beau Flanagan (Jul 2011) tests results: “245 million over 36 domains probably just isn’t enough”.
I was thinking on its uses for counter-like purposes when I found two solutions by Mocky on Sep 2009 for this: dumping data to sql servers after analyzing it (I will give this a try) and a harder solution.
AWS DynamoDB: http://aws.amazon.com/en/dynamodb/
Geourgeous. Fast scalable. Intervention 3 over 10. Take a look to this: Amazon DynamoDB Overview 3 mins video. You have to set a read and write operations limits per second, this will reserve the hardware and requirements for you. You can easily increase or decrease the r/w limits depending on your needs and create alerts to get advised when you be near to reach your limit. The price can be a problem: (US) $0,01 / 10 write ops and $0,01 / 10 read ops. Storage is $1 GB/month. I strongly recommend you to read the dynamodb developers guide, in special how tables and items works. There are things like Consistent Read, Atomic Counters and Conditional Writes you will like to know.
MongoDB: http://www.mongodb.org/
STRONG community support and cool documentation. On its documentation you can find how to setup it on Amazon EC2. And there are drivers for dozens on languages. Its really an option to consider.
You will have to invert more time on setup and scaling and your costs can came more unpredictable when using EC2. Also, if you are thinking on fast growth, a simple sharding architecture will require you setup 3 servers. The community support and documentation are incredible, as commercial support is. You will also find a lot of events and user groups around the world. I have found on forums that many people and startups are using mongo over dynamo on its beginnings keeping in mind the possibility to migrate to dynamo in the future.
See MongoDB use cases.
My last purchase from Amazon.
http://www.amazon.com/The-Design-Business-Competitive-Advantage/dp/1422177807
