Move splunk cloud data to AWS

If the data is stored in reasonably large files, on the order of tens of megabytes in size per file, or if they are smaller but you can bundle multiple files up into zip files or tarballs, for example, then for cold archive storage, the most cost-efficient option is to place them in an S3 bucket and use a lifecycle rule to transition them to the Glacier Deep Archive storage class.

The storage cost for Deep Archive is about $1 per month per terabyte.

Accessing data in Deep Archive requires going through a restore process, which takes between 12 and 48 hours, depending on how much you're willing to pay for the restoration, but since you mentioned the need is specifically for cold storage, it's almost certainly the right option for you.

The reason small files should be combined into larger ones is that there's per-object overhead of several kilobytes in Deep Archive. You can find the details in the price list: https://aws.amazon.com/s3/pricing/