I am creating a data-repository more like creating data-lake for no-SQL db. I have some field which doesn't have a proper schema. They have mix type object like field value is {a:2} or {b:2,c:4, a: {1,2}}, etc.
I can use CSV format so I can save space/storage but due to unstructured/schema-less objects, I am going with the JSON file.
Is there any other way to store data??
I am using AWS S3 as storage for data-lake.
