Creating a dataset - Amazon FinSpace

Creating a dataset

Important

Amazon FinSpace Dataset Browser will be discontinued on March 26, 2025. Starting November 29, 2023, FinSpace will no longer accept the creation of new Dataset Browser environments. Customers using Amazon FinSpace with Managed Kdb Insights will not be affected. For more information, review the FAQ or contact AWS Support to assist with your transition.

Note

In order to create and manage datasets, you must be a superuser or a member of a group with necessary permissions – Create Datasets.

A dataset can be created by loading a file using the Amazon FinSpace web application.

To create a dataset

  1. Sign in to the FinSpace web application. For more information, see Signing in to the Amazon FinSpace web application.

  2. On the left navigation bar of the home page, choose Add Data.

  3. Drag and drop a .csv file or choose Browse Files to select a file. Once the file is detected by the web application, schema of the file will be displayed. The column names are read from the file and data types are inferred.

    A screenshot that shows the Add Data page in FinSpace.

  4. Change the data types as required by choosing Edit Derived Schema. Take note of the data types and formats that are supported.

  5. Choose Save Schema.

  6. Choose Confirm Schema & Upload File. This action starts the following process:

    1. Create a dataset with name of the .csv file that was loaded and takes you to the dataset details page.

    2. Once the upload of the sample data file is complete, a changeset is created with the content of the data file. Verify by checking the Dataset Update History table under All Data Views tab.

    3. Data view creation process is started. Once the upload of the sample data file is complete, a process is kicked off to create a data view that can be analyzed in a notebook.

      For small files of up to 100 megabytes, data view creation takes approximately 2 minutes. For larger files of around 1 gigabyte, expect data view creation to take approximately 3-4 minutes. Views with partitioning and sorting schemes may take longer.

      Once a dataset is created, you can start adding data to it. A new set of data added to a dataset creates a corresponding changeset.