Cassandra excels in a broad range of applications -- especially if you understand its data model and write your applications accordingly. It's an excellent choice for time-series data, and a poor choice for application queues. It performs the best if you can simply record history and compute from it, rather than going back and editing or deleting things a lot.
Perfect solution for caching needs. If you have a bottleneck due to frequent data access to your database, then Redis can really help you by diverting those traffic away from your database. Its key/value pair structure also makes data lookup very efficient, providing excellent performance.
High Availability - we utilize the data replication features of Cassandra. This enables us to access our data even when several nodes have gone down
Data Locality - our architecture combines Cassandra storage nodes and computation nodes in the same machine. This enables us to utilize data locality and limit expensive network IO to read data.
Elasticity - Cassandra is a shared nothing architecture. Nodes can be added very easily and they discover the network topology. As soon as a node has joined the Cassandra ring, the data is redistributed among the existing nodes and streamed to it automatically.
Easy for developers to understand. Unlike Riak, which I've used in the past, it's fast without having to worry about eventual consistency.
Reliable. With a proper multi-node configuration, it can handle failover instantly.
Configurable. We primarily still use Memcache for caching but one of the teams uses Redis for both long-term storage and temporary expiry keys without taking on another external dependency.
Fast. We process tens of thousands of RPS and it doesn't skip a beat.
No Ad-Hoc Queries: Cassandra data storage layer is basically a key-value storage system. This means that you must "model" your data around the queries you want to surface, rather than around the structure of the data itself.
There are no aggregations queries available in Cassandra.
Redis is super fast but it comes with a cost. Whole dataset resides in RAM. So it can be costly as primary memory is more costly, then secondary ones.
Persistence issues: To achieve it, Redis uses a memory dump to create a persistence snapshot, that's cool. But it requires some Linux Kernel tweaking to avoid performance degradation while the Redis server process is forking. This further causes latency.
Master-slave structure side effect: Master-slave architecture comes with its own side effects. Please note that there will be only one master with multiple slaves for replication. All writing goes to the master, which creates more load on the master node. So, when the master goes down, the whole architecture does.
I would recommend Cassandra DB to those who know their use case very well, as well as know how they are going to store and retrieve data. If you need a guarantee in data storage and retrieval, and a DB that can be linearly grown by adding nodes across availability zones and regions, then this is the database you should choose.
We will definitely continue using Redis because: 1. It is free and open source. 2. We already use it in so many applications, it will be hard for us to let go. 3. There isn't another competitive product that we know of that gives a better performance. 4. We never had any major issues with Redis, so no point turning our backs.
It is quite simple to set up for the purpose of managing user sessions in the backend. It can be easily integrated with other products or technologies, such as Spring in Java. If you need to actually display the data stored in Redis in your application this is a bit difficult to understand initially but is possible.
The support team has always been excellent in handling our mostly questions, rarely problems. They are responsive, find the solution and get us moving forward again. I have never had to escalate a case with them. They have always solved our problems in a very timely manner. I highly commend the support team.
Apache Cassandra has the best of both worlds, it is a Java based NoSQL, linearly scalable, best in class tunable performance across different workloads, fault tolerant, distributed, masterless, time series database. We have used both Apache HBase and MongoDB for some use cases which were within hadoop setup and JSON (JavaScript Object Notation) document store respectively, but given the overall factors favoring Apache Cassandra, it is a technology choice for multiple platforms!
UI isn't that great compared to the other competitors. The management of our memcached cluster was becoming pretty complicated as the application grew in size. Redis is a much better option compared to memcached. Redis is bit unreliable compared to the alternative RabbitMQ especially when it needs to be integrated with Celery.
The open source version of Cassandra is only suggested for learning the basic concepts and play with its core features. Unless you really want to invest a lot in your developers and architects knowing every detail of Cassandra, I prefer the DataStax enterprise version. Although the license cost is relatively high, I think they it is worth it. I'm thinking about the support, the monitoring tool OpsCenter, and the integration of Solr and Spark (for data analysis).
Cassandra didn't fully replace our old and traditional relation database Oracle. In addition, it opens another door for us to deal with some special business use cases that NoSQL database can do better in a more feasible and efficient way.
Existing tools like Redisson that were built over Redis reduced dev time in solving challenging problems, which had a positive impact on ROI.
We initially misused Redis for persistent storage which had a negative impact on ROI because we were paying a lot for inactive users.
The increased performance we achieved using Redis in areas like locking helped us improve the performance of our system reducing the likelihood of system timeouts.
