Last week I was on vacation in Texas, bouncing around San Antonio, Austin, Houston and several points in between, and I was lucky enough to be able to attend the MongoDB Road Show on Thursday afternoon. Being in IT for over a decade, I attend these events with a skeptical eye. Go for the Swag, Stay for the Information is my honest opinion. Generally if its good a presentation, I can get something out of it, but if not, at least I got a T-Shirt or some other branded item I can use. I’m happy to report this was a successful trip. I’ve only touched briefly before on MongoDB here, not because I am unfamiliar with it, just that in my day to day work, it doesn’t come up. I made an unsuccessful attempt a couple of years ago to get our development team interested in NoSQL in general and MongoDB specifically, but there was little interest. Despite that, I’ve still paid attention to what’s been going on in this arena, though, and I’m going to attempt to post more on MongoDB in the future. MongoDB is industry leader in the NoSQL world according to the initial presentation. There is a large number of job opportunities for developers familiar with it, a large and growing installed user base, the company is attracting large amounts of investment dollars from various well known venture capital groups, and its one of the most frequently searched for database systems. All of that seems pretty positive, but I wonder is it attracting attention because they have the best product, or is it because its just good enough? MongoDB certainly has one of the best marketing machines out there, far superior to other NoSQL vendors like Hortonworks (Hadoop), DataStax (Cassandra) and CouchBase. And as they gain market share, they have partnered with other vendors to expand the ecosystem.
Pentaho reps were present at the Road Show, providing free copies of a book on integrating their products with MongoDB and presenting on how easy it is to pull data to and from MongoDB. Pentaho Analytics now supports MongoDB across their analytics platform, not just in Kettle. I’ll be testing and writing about that in future entries here.
Another vendor – Pure Storage was also present (giving away tshirts and coozies) and talked about how their flash disk arrays improve disk read/write speed by several orders of magnitude. One of the biggest bottlenecks for large scale data sharding and redundancy is the amount of time it takes to write the data to multiple machines. By using flash storage, they are cutting that latency down significantly. By partnering with them, the MongoDB team is addressing some concerns that have been voiced in the field that data was not synced and verified across multiple nodes before MongoDB moved on.
Some of the other enhancements that were covered indicate that the MongoDB team is actively listening to the user base and working to improve their product. The latest release (MongoDB 2.6.) introduced the MMS – management system that allows you to monitor your cluster from another machine. And some of the biggest roadblocks to larger enterprise adoptions are being worked on and rolled out including enhancements in Security, BI and Analytics, and Text Search. Cramming all of this information into a couple of hours on a hot June afternoon gave me some new insight into MongoDB and the direction they are working towards. While I’m still not sold on them, I am ready to take another look at their product and try it out. More to come….