AWS Timestream - Batch Load Tasks

Thank you, I seemed to have figured it out. I will write what I did here.

1. https://docs.aws.amazon.com/timestream/latest/developerguide/batch-load-prerequisites.html In this link you sent me, Timestream requires that the time is in Unix Epoch time. This was one of the mistakes that I was making, so I converted the time into Unix Epoch time in excel with this formula: (A2-DATE(1970,1,1))*86400. See number 4 below for an additional problem with this.

2. Another mistake was simply, when I renamed my file, it had a space in it (for example: batch load.csv). I renamed it to this instead: (batch-load.csv) this time with no space and a hyphen instead, and it worked.

3. (I think this file-opening process can be ignored as long as you save it as a Comma-Separated-Values (.CSV) file, but I will leave this in here just in case.) Additionally, I was opening my files my double clicking on the CSV file which automatically opened it in excel. However, excel attempts to interpret CSV files and changes it in the process, which causes problems when exporting the file again. Instead, I followed this website which uses a method with a text file to safely open it in excel. Here is the link: https://help.supportability.com.au/article/250-how-to-open-csv-files-safely-with-microsoft-excel Also, for exporting the file again, I just rename the blank excel workbook it the exact same thing that it was called before and save it as Comma Separated Values (.csv).

4. In Timestream on Batch Load Tasks, while filling out the Data Model Mapping for the Visual Builder option, I was leaving it on the default Milliseconds option under the Timestamp time unit. When testing the sample file found here: https://docs.aws.amazon.com/timestream/latest/developerguide/batch-load-using-console.html the data is in milliseconds, while my calculated Unix Epoch time (found using number 1) was in seconds. I switched the Timestamp time unit from the Milliseconds option to the Seconds option.

5. Another problem I had was adding a "measure_name" and "dimension" column. The CSV file data set that I was working with did not have these columns (it just has a "timestamp" and "multi" columns). I cannot just add these through the Visual Builder in Data Model Mapping (see number 4). I must go into the file manually (through number 3) and add a "measure_name" column and "dimension" column. I then uploaded it to my S3 bucket under the same name after saving it with the SAVE AS option (described in number 3).

I am new to AWS Timestream but these solutions seemed to work for my problem. Thank you for your response!