Limitations on QuickSight

(只提供英文版本)

Big Data – Blog by YY

Limitations on QuickSight

In the previous blog, I mentioned how to do data analysis in between QuickSight, data visualization service, and, S3, where the sources coming from. Like other services, this kind of integration has pitfall as well. I am gonna list pitfall regarding integrating QuickSight and S3 for off the wrong route in development.

Files, the source being digested from S3 into QuickSight should be in the compliance.- Every single file MUST less than 1GB
– Total files being specified in manifest file MUST less than 10 GB
– Total number of files being specified in manifest file MUST less than 1000
– For files being stored as csv, up to 200 columns.

Data Type, the following primitive data types are in support.
– Date, JODA date formats
http://www.joda.org/joda-time/apidocs/org/joda/time/format/DateTimeFormat.html
– Decimal, up to 4 decimal places to the right of the decimal point.
example, say, 123.45678, the decimal.
ONLY 123.4567 being popped up in QuickSight.
– String
– Integer

With the quick grasp of these limitations on QuickSight and S3 in between, it will be beneficial for developer to
anticipate all the consequences before ETL.

Posted in ICG日志.