Spark Streaming Pros and Cons

Streaming and handling large datasets in Review collected by and hosted on G2.com.

No, i didn't got issues while using spark streaming Review collected by and hosted on G2.com.

It is very use full tool which can teach you more in depth knowledge of big data Review collected by and hosted on G2.com.

It is a very complicated to understands the functions Review collected by and hosted on G2.com.

Native compatibility with spark and ease of use Review collected by and hosted on G2.com.

Basic knowledge of spark is required to start with Review collected by and hosted on G2.com.

Spark streaming has great ecosystem of spark which is a great advantage and it is scalable and fault tolerant Review collected by and hosted on G2.com.

Nothing much but according to me data visualization can be more enhanced and cost can be considered. Review collected by and hosted on G2.com.

You can use spark streaming to process data from wide varieties of system in a distributed manner Review collected by and hosted on G2.com.

File management and latency issue because of near real processing Review collected by and hosted on G2.com.

Intuitive API and robust runtime. Spark can process continuous streams at high data rates. Easy to scale to multiple tasks/workers and numerous nodes in a cluster, providing the performances required. Many clients and services are available to integrate with external storage, database, and streaming frameworks. Review collected by and hosted on G2.com.

Sometimes is difficult to debug, troubleshoot a pipeline. Installation and configuration are complex. Review collected by and hosted on G2.com.

It can handle large amount of datasets in fairly quick manner. The processing is very fast. Review collected by and hosted on G2.com.

It requires ample amount of training and expertise to use and process. Review collected by and hosted on G2.com.

I love the ease of use and how intuitive it is. Review collected by and hosted on G2.com.

I would love an easier file management system. Review collected by and hosted on G2.com.

Spark docs are saying, it is not atomic but it is near-atomic in the way of committing data but it loses in consistency. In our case, we are using a spark engine to read JSON from Apache kafka and write it to Data lake properties. Best part is streaming application on batch based. Review collected by and hosted on G2.com.

It is not good in processing in-memory data when it auto-commit is disabled. Review collected by and hosted on G2.com.