I have a column that has 19 digits and looks like it exceeds the bidwidth of QS. My rows are getting skipped with error NUMBER_BITWIDTH_TOO_LARGE.
Any workarounds for this ?
Hi @Shivani, welcome to the community.
Numeric data includes integers and decimals. Integers with a data type of INT are negative or positive numbers that don’t have a decimal place. QuickSight doesn’t distinguish between large and small integers. Integers over a value of 9007199254740991 or 2^53 - 1 might not display exactly or correctly in a visual.
Decimals with the data type of Decimal are negative or positive numbers that contain at least one decimal place before or after the decimal point. When you choose Direct Query mode, all non-integer decimal types are marked as Decimal and the underlying engine handles the precision of the datapoint based on the data source’s supported behaviors. For more information on supported data source types, see Supported data types and values.
When you store your dataset in SPICE, you can choose to store your decimal values as fixed or float decimal types. Decimal-fixed data types use the format of decimal (18,4) that allow 18 digits total and up to 4 digits after the decimal point. Decimal-fixed data types are a good choice to conduct exact mathematical operations, but QuickSight rounds the value to the nearest ten thousandth place when the value is ingested into SPICE.
Decimal-float data types provide approximately 16 significant digits of accuracy to a value. The significant digits can be on either side of the decimal point to support numbers with many decimal places and higher numbers at the same time. For example, the Decimal-float data type supports the number 12345.1234567890 or the number 1234567890.12345. If you work with very small numbers that are close to 0, the Decimal-float data type supports up to 15 digits to the right of the decimal point, for example 0.123451234512345. The maximum value that this data type supports is 1.8 * 10^308 to minimize the probability of an overflow error with your data set.
Did this answer your question? If so, please help the community out by marking this answer as “Solution!”
Hi , Thanks for the details. This helps