# Understanding your report versions
<a name="understanding-report-versions"></a>

Amazon updates your Cost and Usage Report at least once a day until your charges are finalized. When you create a report, you can choose to create new report versions or overwrite the existing report version with every update.

Your report files include a .csv file or a collection of .csv files and the manifest file. Your report can also include any additional files that support your data’s integration with Amazon Athena, Amazon Redshift, or Quick.

The following sections describe the file organization and naming conventions based on the report versioning that you choose.

## Cost and Usage Reports delivery timeline
<a name="access-cur-s3-timeline"></a>

During the report period, Amazon delivers a new report and a new manifest file each time Amazon updates the report. Amazon builds on previous reports until the end of the billing period. After the end of the report billing period, Amazon generates a new report with none of the information from the previous report.

## Creating new Cost and Usage Report versions
<a name="keeping-previous-cur"></a>

When you choose to keep your previous Cost and Usage Reports, your Amazon CUR uses the following Amazon S3 organization and naming conventions.

```
<example-report-prefix>/<example-report-name>/yyyymmdd-yyyymmdd/<assemblyId>/<example-report-name>-<file-number>.csv.<zip|gz>
```
+ `report-prefix` = The prefix that you assign to the report.
+ `report-name` = The name that you assign to the report.
+ `yyyymmdd-yyyymmdd` = The range of dates that the report covers. Reports are finalized at the end of the date range.
+ `assemblyId` = An ID that Amazon creates each time that the report is updated.
+ `file-number` = If the update includes a large file, Amazon might split it into multiple files. The `file-number` tracks the different files in an update.
+ `csv` = The format of the report files.
+ `zip` or `gz` = The type of compression applied to the report files.

For example, your report could be delivered as a collection of the following files.

```
<example-report-prefix>/<example-report-name>/20160101-20160131/<123456789>/<example-report-name>-<1>.csv.<zip>
<example-report-prefix>/<example-report-name>/20160101-20160131/<123456789>/<example-report-name>-<2>.csv.<zip>
<example-report-prefix>/<example-report-name>/20160101-20160131/<123456789>/<example-report-name>-<3>.csv.<zip>
<example-report-prefix>/<example-report-name>/20160101-20160131/<123456789>/<example-report-name>-Manifest.json
<example-report-prefix>/<example-report-name>/20160101-20160131/<example-report-name>-Manifest.json
```

Amazon delivers all reports in a report date range to the same `report-prefix/report-name/yyyymmdd-yyyymmdd` folder. Amazon gives each report a unique ID and delivers it to the `assemblyId` subfolder in the date range folder. If the report is too large for a single file, the report is split into multiple files and delivered to the same `assemblyId` folder.

For more information on manifesting files when you keep a previous report, see [Cost and Usage Reports manifest files](#manifest-cur-files)

## Overwriting previous Cost and Usage Reports
<a name="overwrite-previous-cur"></a>

When you choose to overwrite your previous Cost and Usage Reports, your Amazon CUR uses the following Amazon S3 organization and naming conventions.

```
<example-report-prefix>/<example-report-name>/yyyymmdd-yyyymmdd/<example-report-name>-<file-number>.csv.<zip|gz>
```
+ `report-prefix` = The prefix that you assign to the report.
+ `report-name` = The name that you assign to the report.
+ `yyyymmdd-yyyymmdd` = The range of dates that the report covers. Amazon finalizes reports at the end of the date range.
+ 
+ `file-number` = If the update includes a large file, Amazon might split it into multiple files. The `file-number` tracks the different files in an update.
+ `csv` = The format of the report files.
+ `zip` or `gz` = The type of compression applied to the report files.

For example, your report could be delivered as a collection of the following files.

```
<example-report-prefix>/<example-report-name>/yyyymmdd-yyyymmdd/<example-report-name>-<1>.csv.<zip>
<example-report-prefix>/<example-report-name>/yyyymmdd-yyyymmdd/<example-report-name>-<2>.csv.<zip><example-report-prefix>/<example-report-name>/yyyymmdd-yyyymmdd/<example-report-name>-<3>.csv.<zip>
<example-report-prefix>/<example-report-name>/yyyymmdd-yyyymmdd/<example-report-name>-Manifest.json
```

### Athena specifications
<a name="overwrite-athena"></a>

If you chose Athena support when you created your Amazon CUR, the file naming conventions are the same as when you choose to overwrite your Amazon CUR except for the format and compression. Athena Amazon CUR files use `.parquet` instead. For example, your report could be delivered as a collection of the following files.

```
<example-report-prefix>/<example-report-name>/yyyymmdd-yyyymmdd/<example-report-name>.parquet
<example-report-prefix>/<example-report-name>/yyyymmdd-yyyymmdd/<cost_and_usage_data_status>
<example-report-prefix>/<example-report-name>/yyyymmdd-yyyymmdd/<example-report-name>-Manifest.json
<example-report-prefix>/<example-report-name>/yyyymmdd-yyyymmdd/<example-report-name>-create-table.sql
<example-report-prefix>/<example-report-name>/yyyymmdd-yyyymmdd/crawler-cfn.yml
```

### Amazon CloudFormation specifications
<a name="overwrite-cloudformation"></a>

In addition to the Amazon CUR files, Amazon also delivers an Amazon CloudFormation template that you can use to set up an Amazon CloudFormation stack that enables you to query Amazon S3 data using Athena. If you don't want to use the Amazon CloudFormation template, you can use the provided SQL to create your own Athena tables. For more information, see [Querying Cost and Usage Reports using Amazon Athena](cur-query-athena.md).

## Cost and Usage Reports manifest files
<a name="manifest-cur-files"></a>

When Amazon updates Amazon CUR, Amazon also creates and delivers manifest files that you can use for Amazon Athena, Amazon Redshift, or Quick.

Manifest files use the naming conventions, and lists the following:
+ All of the detail columns that are included in the report to date
+ A list of report files if the report was split into multiple files
+ The time period covered by the report, and other information.

```
<example-report-prefix>/<example-report-name>/yyyymmdd-yyyymmdd/<example-report-name>-Manifest.json
<example-report-prefix>/<example-report-name>/yyyymmdd-yyyymmdd/<assemblyId>/<example-report-name>-Manifest.json
<example-report-prefix>/<example-report-name>/<example-report-name>/year=2018/month=12/<example-report-name>-Manifest.json
```

### Creating new Cost and Usage Report versions
<a name="manifest-cur-keeping"></a>

When you keep the previous Cost and Usage Reports, the manifest file is delivered to both the date range folder and the `assemblyId` folder. Each time Amazon creates a new Amazon CUR for a date range, it overwrites the manifest file stored in the date range folder with an updated manifest file. Amazon delivers the same updated manifest file to the `assemblyId` folder along with the files for that update. Manifest files in the `assemblyId` folder aren't overwritten.

### Overwriting the previous Cost and Usage Reports
<a name="manifest-cur-overwrite"></a>

When you overwrite the previous Amazon CUR, the manifest file is delivered to the `month=mm` folder. The manifest file is overwritten along with the report files.

### Amazon Redshift specifications
<a name="manifest-cur-RS"></a>

If you chose the option for Amazon Redshift support in your Amazon CUR, Amazon also creates and delivers a file with the SQL commands that you need to upload your report into Amazon Redshift. You can open the SQL file with a regular text editor. The SQL file uses the following naming convention.

```
<example-report-prefix>/<example-report-name>/yyyymmdd-yyyymmdd/<assemblyId>/<example-report-name>-RedshiftCommands.sql
```

If you use the commands in the `RedshiftCommands` file, you don't need to open the `RedshiftManifest` file.

**Important**  
The `manifest` file determines which report files the `copy` command in the `RedshiftCommands` file uploads. Deleting or removing the `manifest` file breaks the copy command in the `RedshiftCommands` file.

### Amazon Athena specifications
<a name="manifest-cur-Athena"></a>

If you chose the option for Amazon Athena support in your Amazon CUR, Amazon also creates and delivers multiple files to help set up all of the resources that you need. Amazon delivers a Amazon CloudFormation template, a SQL file with the SQL to create your Athena table manually, and a file with the SQL to check your Amazon CUR refresh status. These files use the following naming conventions.

```
<example-report-prefix>/<example-report-name>/<example-report-name>/yyyymmdd-yyyymmdd/crawler-cfn.yml
<example-report-prefix>/<example-report-name>/<example-report-name>/yyyymmdd-yyyymmdd/<example-report-name>-create-table.sql
<example-report-prefix>/<example-report-name>/<example-report-name>/yyyymmdd-yyyymmdd/<cost_and_usage_data_status>
```