Apache Cassandra Reviews

It is designed to handle large amounts of data and can scale to support thousands of nodes across multiple data centers. And of course, it is open source which is a plus.

It is a complex database system that requires a high level of expertise to set up and maintain. It can be difficult to configure and optimize.

It is really beneficial for faster write operations and fast read search type queries that can be used in a pipeline with a high velocity of Read/Write operations. Management of data is carried out well by Cassandra and provides replication with its replication factor and replication strategy to prevent data loss.

Apache Cassandra is not meant for analytical queries and it cannot perform certain aggregation functions, Joins like other databases can.

Basically, if you need a database software, you can use it, but you need to be a novice in this business to be extremely satisfied. It needs more improvement and the price should be a little cheaper.

Its overall quality should be improved, Simple shortcuts and special complications added. Its price and payment methods need to be updated, it needs more attention but once the minor glitches are cleared I have no doubt you will release a great piece of software.

First of all, I am a fan of Open Source projects and I really enjoy working with Apache products. I have used Cassandra for several months for my research project that included storing unstructured data. I found it a little bit more difficult to use than Amazon DynamoDB, but given that it is the open source project I was thoroughly impressed by all the features and the reliability of Cassandra. I would recommend Cassandra for projects using Open Source technologies. It is very fast and met all of the efficiency metrics and agreements for the project I have been developing.

I found it a little bit harder to work with compared to MongoDB and DynamoDB. Nevertheless I believe that for an experienced engineer who has worked with Databases before, it will take couple of weeks to get fully fluent with Cassandra docs.

Apache Cassandra is good for bigdata systems.

Apache Cassandra has not translaction support

Easy to download, install, deploy and use.

Need heavy efforts for training new users.

Opensource license provides an easy cost entry point Programmatic support for our purpose was very good.

There is only one commercial version of this product and licensing is high. It can consume far more resources to handle the workload you give it than spec data suggests

- High availability as no master slave concept - Partition tolerance of the system - Replication across multi data centers

- Upgrade to a higher version is non-trivial and requires a lot of testing and evaluation - Cqlsh query results not formatted well which creates confusion in reading

As a user I would like it handle big data workflows across multiple nodes without any single point of failure .and it’s acting like a distributor among all the clusters

It have a capability when nodes goes down read and write request will be stored in other node with in network

The is awesome tool for No SQL DB requirements. We use this to store transactions so that we can reprocess in case something goes down.

The Queries are little tough and most of the features of sql queries are not supported.

The query language cql is quite intuitive specially for sql users. Setup is also very easy since they have extensive documentation. Cassandra also provides very good and easy to use APIs that can be easily integrated in any application.

It is a column oriented storage format which takes some time to get used specially for sql users.

I like it scales very easily when we see increase it traffic

Cassandra does not support joins so it is difficult to use as complex OLTP system

CASSANDRA is easy to use, easy to search objects based on key space values

Cassandra needs to improve their architecture and also provide more training videos

Cassandra is good choice from NoSql database when you need to store data which is coming at high velocity. As write operation in cassandra is very fast so it make it reliable to use it for real time scenarios where you have to process and modify the data in real time. And as Cassandra is built using Java, so it is easy to integrate this with other Java base tools like Spark, Hadoop etc.

Only cons for me is that it’s unpredictable performance because cassandra has lots of asynchronous jobs and other background job which are not in control. In such cases, troubleshooting some issue is big problem.

As everything is designed for a specific problem to solve. Cassandra has it's own use cases where it perform very much reliably and efficiently. This is good when you have to capture high velocity traffic and need to do operation over it. Also, it offers other lots of great like very high writing speed, easy integration with other JVM based service/tool like spark, hadoop etc.

it has a quite bit hard of learning as you will have to setup the cluster for maintenance and all. And Cassandra is specific to use cases you just can't use it for other.

Compared to MongoDB, another NoSQL database, Cassandra's commands are more closely related to SQL and are consequently more approachable to SQL experts. As a NoSQL database, developers are not weighed down with designing and modifying schemas.

Compared to MongoDB, Cassandra is less flexible and resembles relational databases more than MongoDB does.

It is very easy to get started with Cassandra because of the vast community of developers and resources available. It is flexible for any course project or production application

Configuring it to work with Tableau on a Macbook Pro is not that easy.

csql syntax is powerful lightning fast read/writes smart client to reduce hops to the correct node easy to install

cluster management complexity problematic node recovery

It was the best NoSQL solution for our organization. decent performance and cross-Datacenter replication.

It's written in JAVA. GC was killing us. at some point we had a server on GC for 30 min every 15 min. we had to add more and more servers to deal with the situation. DataStax's PHP driver was using the first server available as a 'Contact point' when it was on GC, the second server on the list was hit and sometimes caused to crash; as result - we had a domino effect and an entire cluster would have crashed. (that, btw, was fixed and now Contact point is random). There is a better alternative ***SCYLLADB**** Since we moved to Scylla - we were able to reduce the amount of servers by 75%. Performance and stability is magnificent. client side hasn't changed - running with the same drivers.

It has been a pleasant and profitable experience for the handling of data in the cloud and totally free, the handling of the data oriented to column families, tolerant to failures, since it replicates the data automatically to multiple nodes; When a node fails it can be replaced without any period of inactivity. allows replicas to multiple data centers; data storage type column family. The latest version of CQL3 that made the instructions similar to SQL with what breaks with the mental shock at the time of programming. It really does an efficient handling of data in the cloud.

The fact of having the data stored column by column allows us to return the rows more quickly, but when inserting, updating or deleting a record, it must be done in more than one location; for this reason, this type of database is not recommended for OLTP-type, transaction-oriented and high-concurrency systems.

An easy way to share data across multiple clusters. A proper columnar data store, with some novel quorum modes.

No matter how it's configured it's impossible to have really durable data. Having fault tolerant synchronous modes is useless if a node can die because it ran out of journal space. This should never happen! If a node can't write it's journal it should freeze, not the loose consistency of its local data ( I'm talking about corruption here not the loss of consistency with the other nodes). Columnar formatting and retrieval are actually really inefficient. Data formats in anything other than simple storage are bloated and inefficient.

(1) It is highly available , also, replication around nodes is super fast (2) It is fault tolerant inherently by design (3) It is a distributed database, suitable for handling huge volumnes

Row level inconsistency sometimes poses problems with usage in Cassandra. This means when the same row is inserted and updated at the same time, there can be consistency issues with it.

It is great for the right workload; with heavy writes and eventual consistency, it is a leader in the NoSQL space. I recommend Datastax for support using and supporting it.

Maintaining a cluster and an ops center server (for backups and monitoring) can get quite pricey. The learning curve of maintaining a cluster is a bit long.

No pros were added to this review

No cons were added to this review

Undoubtedly performance is an important reason We have not encountered a single point of failure Scalability of Cassandra is good which is the most important for the companies where demand is scaling day by day.

Cassandra has a wide range of asynchronous jobs and background tasks that are not scheduled by the client, the execution can be eccentric. Because Cassandra is a key-value store, doing things like SUM, MIN, MAX, AVG and other aggregations are incredibly resource intensive if even possible to accomplish. I think querying options for retrieving data is very limited.