real-time Kinesis to Parquet in minutes

Kinesis

Amazon Kinesis is a managed service offered by Amazon that aims to make it easy to collect, process, and analyze streaming data.

Kinesis is a popular choice for streaming backbones because it’s in the AWS ecostystem and relatively simple to set up compared to other data streaming options.

Parquet

Apache Parquet is an open-source, column-oriented data storage format of the Hadoop ecosystem designed to provide fast querying on large datasets. Parquet is routinely used for creating very highly scaled data lakes that can still be queried. Parquet is similar to other column-storage file formats that are available in Hadoop.

Estuary builds free, open-source connectors to extract data from Kinesis with millisecond latency, allowing you to replicate that data to various systems for both analytical and operational purposes. Kinesis data exists in a stream and often benefits from being organized into a data lake or warehosue for analytics with history.

Data can then be directed to Parquet using materializations that are also open-source. Connectors have the ability to push data as quicikly as a destination will handle. Parquet likes files that are around 1 GB each. So, if you have high data volumes, Flow can keep your data lake up-to-date in near real-time.

Talk to Estuary Today

Contact Us

Estuary helps move data from

Kinesis to Parquet in minutes with millisecond latency.

Estuary helps move data fromKinesis to Parquet in minutes with millisecond latency.

About Estuary

Estuary enables the first fully managed ELT service that combines both millisecond-latency and point-and-click simplicity. Flow empowers customers to analyze and act on both historical and real-time data across their analytics and operational systems for a truly unified and up-to-date view.

Flow is developed in the open and utilizes open source connectors that are compatible with a community standard. By making connectors interchangeable with other systems, the Estuary team hopes to expand the ecosystem for everyone’s benefit, empowering organizations of all sizes to build frictionless data pipelines, regardless of their existing data stack.