Using MongoDB (or not) for great science

Obviously a number of organizations have successfully deployed NoSQL data stores. Now most of the time we hear about their successes, how it solved all of their scalability problems. However, not everything goes the way you plan.

In his first post, a blogger by the name of Stavros, describes his experience using NoSQL data stores as backing storage for an experiment. After trying CouchDB and discovering it was not a fit, he transitioned to using a MongoDB. However, his experience with MongoDB was not a good one.

His second post is a follow up to the first post. Stavros discovered that he was using a developmental version of MongoDB. After a commenter of his blog pointed this out to him, he made the change to use a stable version. However, unfortunately for Stavros his issues persisted. To sum up his feelings for NoSQL data stores:

It goes without saying, of course, that I will not be touching MongoDB again with a ten-foot pole.

Now this is not an attack on MongoDB obviously there are a number of factors that could be in play. However, it's important that folks see the other side of the fence when deployments don't go as well as you plan.

Read More:

My experience with using MongoDB for great science
Using MongoDB for great science, part 2