Regarding cost optimizations what would be the cheapest

aws s3 cp local-data  s3://same-backet/latest
aws s3 cp local-data  s3://same-backet/some-date

or

aws s3 cp local-data  s3://same-backet/latest
aws s3 cp s3://same-backet/latest  s3://same-backet/some-date

I.e. does bucket-to-bucket copy would use less external (to AWS) traffic then local-to-bucket or it would it make even bucket-local-bucket trip to complete the task?

CodePudding user response：

Here are few insights using which you can determine the cost that would be incurred for S3 use.

1. Data transferred between S3 buckets in the same AWS Region does not incur any cost
2. Any kind of data transfer IN S3 does not have any cost
3. Data transfer OUT to internet is free upto 1GB/Month

So if you transfer data from your local machine to S3, that will be IN, so not cost, safe to use the below transfers

aws s3 cp local-data  s3://same-backet/latest
aws s3 cp local-data  s3://same-backet/some-date

Your second option is also safe as you are transferring data in the same bucket, can also transfer data to another bucket in the same region without having to spend extra.

Suggesting to review the AWS S3 pricing page.

CodePudding user response：

All the answers by @subhashis-pandey are correct but S3 also cost for GET and PUT objects inspite of object(s) being big or small, if your data can be ziped to single file, you need to keep ziped version of data where ever possible.