Meta Broadcast

we write about the things we build and the things we consume

atlas: the cassandra odyssey continues

If you read our blog regularly, and even if you don't, you probably know by now that we're in the process of moving from MongoDB to Cassandra as our primary database behind Atlas. So, as Mirona previewed a few weeks ago, we took a trip to the bright lights of Camden Town to share our experiences with the London-based Cassandra community.

how it all unfolded

The evening kicked-off with a beer or two provided by our hosts, Forward, and plenty of tasty pizza. I started off our presentation with a little scene-setting and an overview of Atlas and its data model. I also spoke about our time with MongoDB including why it was chosen in the first instance, what we like about it and also where we feel it falls short for Atlas.

Tom moved onto to how we are introducing Cassandra into our infrastructure and the intricacies involved such as custom secondary indexes and adding elasticsearch into the mix.

To finish up, Adam ran through his experiences of maintaining and monitoring a healthy Cassandra cluster from a dev-ops perspective including token management, DataStax's supremely useful OpsCenter and the Joy Of Disaster Recovery.

If you want to flick through the slides, here you are:

After a few questions the general consensus was that Adam was the most interesting man in the room and he quickly grew quite the fan-club, sharing stories with the folk from Acunu. I think we all had some illuminating conversations with people with similar tales to us.

Overall I think it's safe to say that it was a pretty good experience and we were able to confirm that we are heading in the right direction. Hopefully everyone there enjoyed it as much as we did. All that remains to be said is a big thank you to all at the Cassandra Meetup and its sponsors: Forward, DataStax and Acunu. We look forward to seeing you there next time.

categories

blog comments powered by Disqus
metabroadcast

flickr