We compared Cassandra and MongoDB based on our users' reviews across five parameters. After reading all of the collected data, you can find our conclusion below.
Cassandra offers scalability, high availability, fault tolerance, and distributed architecture, with praised customer service. MongoDB highlights flexibility, scalability, advanced query language, and replication, with a need for improved query language, documentation, performance, and integration. MongoDB also offers flexible pricing and strong ROI, with exceptional customer service and support.
Features: Cassandra's valuable features include scalability, high availability, fault tolerance, and distributed architecture. Users praise its ability to handle large data volumes and seamless performance across multiple nodes. MongoDB offers flexibility, dynamic data handling, scalability, advanced query language, and reliable replication.
Pricing and ROI: The setup cost for Cassandra is considered straightforward and easy to manage, with flexible and accommodating licensing options. MongoDB offers a seamless experience with a transparent pricing structure and hassle-free setup. Users appreciate MongoDB's various pricing plans and the accessibility of its open-source community edition. Both Cassandra and MongoDB offer a positive ROI. However, Cassandra focuses on improved efficiency, scalability, and performance, while MongoDB is praised for its flexibility, ease of use, and ability to handle large amounts of data efficiently.
Room for Improvement: Cassandra can benefit from improvements in scalability, performance, support for large datasets, handling of write-heavy workloads, ease of use, management of clusters, documentation, and monitoring tools. MongoDB needs enhancements in terms of its interface, navigation system, performance, scalability, documentation, customer support, and data replication capabilities.
Deployment: Some users find the setup process straightforward and easy. They mention that there is documentation and assistance available, which makes the installation process easier. However, other users mention that the initial setup can be difficult and may require assistance. The duration required for the initial setup, deployment, or implementation phases of MongoDB also varies. Some users found the initial setup to be quite easy and straightforward, taking only a couple of hours or even less than an hour. However, there were also users who found it to be a bit complex, especially for specific use cases or cluster deployments, which took a couple of days.
Customer support: Users highly praise the customer service for both Cassandra and MongoDB. They appreciate the promptness, reliability, and knowledge displayed by the support teams. Additionally, the staff's helpfulness, responsiveness, and expertise is appreciated.
The summary above is based on interviews we conducted recently with Cassandra and MongoDB users. To access the review's full transcripts, download our report.
"I am getting much better performance than relational databases."
"The most valuable features of this solution are its speed and distributed nature."
"Cassandra has some features that are more useful for specific use cases where you have time series where you have huge amounts of writes. That should be quick, but not specifically the reads. We needed to have quicker reads and writes and this is why we are using Cassandra right now."
"The most valuable features of Cassandra are the NoSQL database, high performance, and zero-copy streaming."
"The most valuable features are the counter features and the NoSQL schema. It also has good scalability. You can scale Cassandra to any finite level."
"Cassandra is good. It's better than CouchDB, and we are using it in parallel with CouchDB. Cassandra looks better and is more user-friendly."
"A consistent solution."
"Some of the valued features of this solution are it has good performance and failover."
"The aggregation framework is very powerful when elaborating on data."
"Easier to maintain the data with its document-based storage."
"It stores historical data with ease. For example, if you are a healthcare member, then you will have multiple records of visits to the doctors. To store such data in Oracle Database, you have to create many records. You might also have duplication problems because your records are going in again and again, because of which the data warehouse and the maintenance cost will be huge. MongoDB is comparatively lightweight. It is a JSON extract. Once you define a schema and extract it, you can push all the relationships in any way you want. It is easier to define and get different types of transactions into MongoDB. It is also easier to set it up as compared to other solutions. MongoDB is a NoSQL database, which means it is a document DB in which you can store documents that you created in BSON. It is pretty fast in response. It is faster than relational databases because it does not define any primary keys, secondary keys, tertiary keys, and all those kinds of things."
"like its performance and the stability. It's very stable and, performance-wise, it's really great."
"The solution has good flexibility and very fast performance for searching data."
"Sharding is an excellent feature of MongoDB."
"One of the first things I noticed when I had my first experience with MongoDB was how easy it was to use. I was expecting more difficulties or at least some challenges, but it was very, very easy to use. It's great technology, performs well, and is very convenient."
"I like the schemaless architecture that it follows. I also like the sharding that it provides."
"The solution doesn't have joins between tables so you need other tools for that."
"Maybe they can improve their performance in data fetching from a high volume of data sets."
"There were challenges with the query language and the development interface. The query language, in particular, could be improved for better optimization. These challenges were encountered while using the Java SDK."
"Depending upon our schema, we can't make ORDER BY or GROUP BY clauses in the product."
"Doesn't support a solution that can give aggregation."
"Cassandra could be more user-friendly like MongoDB."
"The solution is limited to a linear performance."
"The secondary index in Cassandra was a bit problematic and could be improved."
"They could provide more documentation and examples for adding pipeline stages."
"I'd like to see an ID generator. It's very technical but I don't think it has one, so we have to go to great lengths to work around that."
"I would like to see the scalability and security improved."
"The scalability of the solution has room for improvement."
"MongoDB can improve large-size video or media frame operations. There are a lot of customers who want to upload media frames and video games but there is some difficulty. In MongoDB, we are looking out for solutions that are for large-size media files that can be saved and navigated efficiently."
"Lacks sufficient scalability and elasticity."
"From my point of view, they need a totally free IDE to work at high levels."
"MongoDB should be more stable, and support should be more efficient."
Cassandra is ranked 4th in NoSQL Databases with 19 reviews while MongoDB is ranked 1st in NoSQL Databases with 69 reviews. Cassandra is rated 8.0, while MongoDB is rated 8.2. The top reviewer of Cassandra writes "Well-equipped to handle a massive influx of data and billions of requests". On the other hand, the top reviewer of MongoDB writes "Lightweight with good flexibility and very fast performance for searching data". Cassandra is most compared with Couchbase, InfluxDB, ScyllaDB, Oracle NoSQL and DataStax, whereas MongoDB is most compared with InfluxDB, Couchbase, ScyllaDB, Oracle NoSQL and Oracle Berkeley DB. See our Cassandra vs. MongoDB report.
See our list of best NoSQL Databases vendors.
We monitor all NoSQL Databases reviews to prevent fraudulent reviews and keep review quality high. We do not post reviews by company employees or direct competitors. We validate each review for authenticity via cross-reference with LinkedIn, and personal follow-up with the reviewer when necessary.
SQreamDB is a GPU DB. It is not suitable for real-time oltp of course.
Cassandra is best suited for OLTP database use cases, when you need a scalable database (instead of SQL server, Postgres)
SQream is a GPU database suited for OLAP purposes. It's the best suite for a very large data warehouse, very large queries needed mass parallel activity since GPU is great in massive parallel workload.
Also, SQream is quite cheap since we need only one server with a GPU card, the best GPU card the better since we will have more CPU activity. It's only for a very big data warehouse, not for small ones.
Your best DB for 40+ TB is Apache Spark, Drill and the Hadoop stack, in the cloud.
Use the public cloud provider's elastic store (S3, Azure BLOB, google drive) and then stand up Apache Spark on a cluster sized to run your queries within 20 minutes. Based on my experience (Azure BLOB store, Databricks, PySpark) you may need around 500 32GB nodes for reading 40 TB of data.
Costs can be contained by running your own clusters but Databricks manage clusters for you.
I would recommend optimizing your 40TB data store into the Databricks delta format after an initial parse.
Morten, the most popular comparisons of SQream can be found here: www.itcentralstation.com
The top ones include Cassandra, MemSQL, MongoDB, and Vertica.