Amazon Kinesis Data Streams Reviews & Product Details - Page 7

Easy to manage as compared to Kafka streaming Review collected by and hosted on G2.com.

Cost as compared to open source Kafka and difficult in understanding Review collected by and hosted on G2.com.

I have been using Amazon Kinesis from a long time for generating real time data from multiplayer games and video streamings,etc. The data generated can be shared with another applications. Kinesis produces data in milliseconds. Review collected by and hosted on G2.com.

Amazon kinesis requires lot of storage space as it generates high amount of real time data. Apart from this, i have not faced any issue with kinesis. Its a trustworthy product by Amazon. Review collected by and hosted on G2.com.

Batch operations and tumbling window processing of data. Review collected by and hosted on G2.com.

Documentation sometimes feels lacking in clarification between data streams, kinesis analytics etc Review collected by and hosted on G2.com.

We like the serverless facility of Kinesis so much, Kinesis can handle any kind of real-time streaming data in real-time.Great application for real-time Machine Learning applications Review collected by and hosted on G2.com.

Still, Kinesis should support a lot of ETL processes, similar to AWS glue. And looking for more sample project of getting started with Kinesis for IoT Review collected by and hosted on G2.com.

Best service to capture events from the dynamo streams Review collected by and hosted on G2.com.

Very difficult to find out the missing events when the events flows on to headwaters Review collected by and hosted on G2.com.

It's extremely simple to set up and you don't need to build out the typical Kafka infrastructure. It scales really well and is reliable. Review collected by and hosted on G2.com.

Some of the advanced features, such as sharding and other horizontal-scaling issues, can be hard to do, but once you get comfortable with event-driven paradigms, it all will make sense. Review collected by and hosted on G2.com.

I didn't have to setup anything and by just wrting consumer our streaming service was ready to go. Review collected by and hosted on G2.com.

I understand by design it is at-least once architecture, but the number of duplicate records we get is not less. On regular basis we see at least few records coming in as duplicate. To avoid duplicate record processing we had to write ourown login and cache the records in redis.

We use KCL to consume the data and some of the default confuration can lead to data loss. We had to checkpoint every record and it seems by default checkpoint also connects with Kinesis that results in more than 5 connection to Kinsis within seconds resulting in error. We eneded up lossing data in production because of this issue Review collected by and hosted on G2.com.

I like how easy it is with AWS Kinesis to stream your data and hook it up with Lambda functions that can process that data afterwards. Review collected by and hosted on G2.com.

If you have a lot of different data entities that you need to stream, or if you have big batches of data, you have to use more shards. So it immediately becomes a very expensive service to use. Review collected by and hosted on G2.com.

Ease of use. From the time we decided to use Kinesis to the time I had a POC done was about 3 hours. Review collected by and hosted on G2.com.

The producer library and the consumer library were obviously written by different teams. Instantiating a producer is way different than consumers. Review collected by and hosted on G2.com.

The best thing about kinesis is we can pass on the data to any other service on aws through kinesis firehose. Its the SaaS version of Apache kafka. With very high throughput distributed. We can use it as a pub sub system as well. Review collected by and hosted on G2.com.

I dislike that we can not make small query data in to a stream directly for that we have to use Kinesis data analytics and that is an extra thing to do. Review collected by and hosted on G2.com.