MongoDB has been known for a while for its flexibility and performance scalability. Come version 2.8 of MongoDB, the whole notion of flexibility, scalability and concurrency is going to be taken to newer heights.
Let’s take you through what to expect in the upcoming version 2.8 of MongoDB, which is probably a milestone version in its history so far.
1. Locking and Concurrency –
Locking and concurrency are two sides of the same coin. The granular you can keep locking on resources the more locks you can acquire and hence more will be the number of users you can support. So basically the scope of lock would decide how many concurrent operations can happen. Prior to v2.2 of MongoDB this scope was global, which meant a database db1 would be locked even if the database being operated upon is db2. It got better with v2.2, where the scope was narrowed down to database. Still it posed concurrency challenges for systems where there is huge simultaneous write requests. With version 2.8 you will be able to experience document level locking. This is a great news, especially for high-write workloads. So like I said, with narrower scope of locking we are up for more scalable MongoDB implementations.
2. Pluggable Storage Engine –
Among the rise of so many NoSQL technologies, MongoDB’s true advantage has been versatility. MongoDB claims to fit into more NoSQL scenarios than any other NoSQL store. Having said that, MongoDB has been deployed in vast scenarios and different workloads. Though programmatically, it is scalable and can fit into all those spaces, it has been missing its operational counterpart in its endeavour to achieve scalability. With v2.8, you will have the facility to have a pluggable storage API that allows developers to choose a storage engine of their choice. What does that mean to you? As Eliot declared in his keynote at MongoDB World, “You could actually have a replica set with multiple storage engines: one optimized for BI workloads, one focused on compression, etc.”
Now this completes the MongoDB’s versatility claim to fit into various workloads.
3. Automation –
Since its inception, MongoDB has been focused on making life easier for ‘DevOps’ (DBAs as well as Developers). That has always been the case. Be it with transparent Auto-Sharding which makes life easier for developers to write codes for a distributed database. Or be it with a replica set auto-healing/auto-failover/auto-join for replicas, which makes life easier for DBAs and operations team – Well I never liked those wake-up calls in the middle of a night to do a failover and rebuild the lost replicas. The MMS backup has made PITR and backup recovery as easy as a walk in the park. The automation has been made more lucrative with addition of MMS automation. It will take the provisioning and upgrades for MongoDB database to an all new lows. Yes, you heard it right! All new lows of difficulty!
As MongoDB adds more features and rides up the scalability ladder, we at Ashnik are watching it closely. We are now engaged in few evaluations and POCs around MongoDB, while seeing a definite excitement and appetite in the market. Both MongoDB community and MongoDB Inc. have been very responsive to this demand. We were present at the MongoDB World earlier this year and it was a big and true representation of its popularity which is now evident in this year’s Gartner Quadrant of Operational Databases. Stay tuned, as we bring in more about this great NoSQL database to your mailbox in our future newsletters. Meanwhile, if you need for NoSQL store do drop us a note.