Hi All,
In this post I am going to talk about a very unique NoSQL database called Cassandra. Which situation is this best suitable for and where to avoid using it.
About Cassandra: Cassandra is a free open source NoSQL database. It was created by Facebook and then open sourced to Apache. It is a distributed, wide column storage system designed to handle big data.
It provides high availability and scale to it with no single point of failure.
When I first decided to use Cassandra there were couple of reasons behind choosing this as a database provider. I had a very large set of data coming to me which needs to be stored and analysed as well. I needed a fast read and fast write throughput.
I did a pros and cons analysis of it which i want to share with everyone now as i am using it for a while now.
Pros:
- It has both SQL and NoSQL capabilities
- Read and Write both are fast
- Can handle big dataset very easily
Cons:
- Very less support ecosystem
- Primary keys are fixed so it has a very rigid table structure, so need to finalise the primary keys before hand only
- Modelling of tables is very important and rigid
- Very complex query system
- Select__IN query is big NO
Conclusion:
Cassandra is ideal for big data where you need good throughput and good analysis. It is best for
- IoT devices data
- Storing the chat message.
- Where relations and a unique primary id is not necessary and only primary keys are required to insert and get queries
If you are interested to implement Cassandra as your database system, will cover that in a separated blog.
