MongoDB Berlin 2013

Today, the MongoDB Berlin 2013 took place at the bcc. Like in the past years there were parallel tracks with lots of talks. Here are the ones I've attended:

  • MongoDB Internals (How does Mongo store my data?): Covered the directory layout and some tuning options, e.g. concerning the preallocation strategy, the internal file format, the extent structure and indexes. Also covered: memory mapped files and their pros & cons, journaling incl. tuning and fragmentation, e.g. due to remove() and update(), and strategies to work around this. Storage-viz is a little tool to help understanding the MongoDB storage layout.
  • JSON All Over (Efficient web development with MongoDB and Node.js): Presented the advantages and characteristics of JSON, JavaScript and Node.js. The presentation itself  was a Node.js app with data stored in MongoDB.
  • Bringing Some Spatial Love to Your Application: An OpenShift (a PaaS product) promotion and a little demo concerning MongoDB spatial indexing and geolocation.
  • MongoDB and Node.js (An introduction to the node.js mongo driver): Introduction into Node.js, event-driven development and the corresponding MongoDB driver (covering cursors, streams, GridFS, BSON up to the client aka "raw mode") with some examples and an outlook to some experimental stuff. Mentioned Learn MongoDB And Node.js The Hard Way.
  • Optimising for Performance, Scale and Analytics: Covered some EC2 performance tips, bottlenecks, the benefits of covered indexes, the hint to drop collections/databases instead of bunches of documents and monitoring tools like mongostat. Here is the blog of the speaker with MongoDB-related posts.
  • Sharding - Hands On: Held by one of the organizers of the MongoDB User Group Berlin. Covered the basics on sharding and introduced an example scenario incl. demo. The files to play around with sharding can be found here.

No comments:

Post a Comment