Cassandra Reviews

I designed a product catalog data model in Cassandra according to their features and properties, loading millions of data and performing the required queries over it. 

Now, I am getting much better performance than relational databases.

Here are some features which have been really helpful in my organisation:

• High Avaialability
• No downtime
• Better performance
• Greater scalability.

HA is one of the great features of Cassandra with no downtime, e.g., you can achieve continuous data without a single downtime because of node to node ring architecture.

Maybe they can improve their performance in data fetching from a high volume of data sets. 

I had to compare Cassandra with MongoDB. MongoDB is much better in data fetching than Cassandra.

We've used Cassandra in the past to design a right-node read-less ideology. We mainly use it for its database capabilities.

Right now, the solution is working very well.

Cassandra has a very good understanding of GBL, and how to cure GBL in time. The biggest problem is always with GBL in terms of understanding the drives' collector and making them work perfectly. Cassandra addresses this very well.

The solution's database capabilities are very good.

We actually find HBase to be faster and better than Cassandra.

The disc space is lacking. You need to free it up as you are working.

I have about ten years of experience working with the solution.

We have some experience with HBase, which we find to be a faster solution.

My first Cassandra project was with a project introduced to us by Facebook. That was ten years ago. There was a time I tried using it a couple of months ago, and I completed the project for Upwork for Cassandra. Right now, I have another project which is using a Cassandra cluster which is under my management. Previously, I had quite a big Cassandra cluster of about 100 nodes and about 500 terabytes of data.

Overall, I would rate the solution nine out of ten.

I really appreciate the high availability, automated replication, linear scalability, and automated region fail-over.

We've used Apache Cassandra for solutions that we sell to our customers. It's used as our cloud based backend store as a temporary cache and for storing data that streams through our data pipe. It's an excellent high speed store.

Out-of-the-box monitoring, troubleshooting, and maintenance are involved. There are several utilities/interfaces available for use, but one would have to educate himself and learn the intricacies of managing a Cassandra cluster.

For example, we recently hired a consulting firm to make recommendations on how to approach maintenance and the health of the cluster and we're learning from that experience.

I have used this product for approximately two years.

We have had stability issues including out of memory issues and crashes with earlier versions of the product.

We’ve not really had scalability issues, but scalability is solved by advanced tuning or adding nodes.

We use the open source version, so support is pretty much on our group of developers and public forums/user groups. For example, I'm a member of the Cassandra user list mail group.

This is a new cloud based enterprise product, so there weren't previous solutions.

The setup is not terribly complex, but a learning process was involved.

We use the open source version, so it's free. Costing needs to take into account home grown maintenance and support, as that can get involved.

We looked at Hadoop, Spark, Spark Streaming, and MongoDB.

If you plan to use the open source version, make sure you hire a Cassandra expert or train yourself in the internals of Cassandra.

• Simple ring architecture
• Replication using peer-to-peer protocol rather than leader selection
• Linear scalability (99%)
• Configurable consistency

Ability to achieve write speeds 10k tps: Compared to existing, it is 300% percent higher.

Row-level locking is not available; might be very helpful in update use cases.

for the past 2 years

older version of Cassandra 2.x having problem while restarting the nodes to the ring. New version from Cassandra 3.5.x onwards this issue got resolved. We are free to stop and start any nodes without any issues.

no

no

The earlier solution used Couchbase, which has leader selection. At times, when leader selection takes time, then we would lose the transactions. This got resolved with the peer-to-peer architecture solution in Cassandra.

Initial setup is straightforward. If you want to do mass cluster setup then centralized tool will be of great help.

in house implementation

Our primary use case is developing software for others and it's really a solution for enterprise size companies. We're like integrators and we have numerous technical partners that implement. We have a partnership with the company, implementing the service on projects. I'm a managing director of the company. 

The most valuable feature for us is the technical evaluation, it's the best technology. Cassandra is good for us. 

The interface could definitely be improved. It's a technical database and for me the features are not user friendly. I also think it's quite an expensive solution and I hope over time with more implementations, this will improve.

I've been using this solution for two years. 

This solution is stable. 

This is a scalable solution. 

From what I know, customer support is fine. 

The initial setup is a little complex and each time we use a specialist for deployment. It depends on the nature of the implementation as to how long deployment takes. 

We don't use this solution like a common database. It's really for people using big data, BI and other analytic software. You need to have the right use case to take this product. 

I would rate this solution a nine out of 10.