MSSQL

MSSQL (Microsoft SQL Server) database S3 Management

Why use S3 for Imports/Exports?

Using S3 allows traffic to flow directly from the RDS instance (where the db runs) to S3 (where the data is stored) without needing to move through the GSA network. This allows data to move much faster and avoid taxing GSA bandwidth for heavy data migration efforts.

Setup

Submit a JSM ticket to get an S3 bucket created per environment for the purpose of holding the database backups. An IAM role will also be created and attached to your db so that it can reach the bucket. Finally, if your db is on-prem an S3 credential and bucket name will be provided for you to faciliate the upload.

Export/backup data to S3

Follow the directions in https://docs.aws.amazon.com/AmazonRDS/latest/UserGuide/SQLServer.Procedural.Importing.html to perform an export/backup of your on-prem or source db.

Uploading a large artifact to S3 potentially requires some customization. Generally speaking an S3 cp command will perform multipart uploads or large datasets. If you have the AWS CLI installed all you need to do is follow instructions in https://aws.amazon.com/premiumsupport/knowledge-center/s3-upload-large-files/. Depending on the machine doing the uploading various parameters may need to be fine-tuned.

Import/restore data from S3

Follow the directions in https://docs.aws.amazon.com/AmazonRDS/latest/UserGuide/SQLServer.Procedural.Importing.html#SQLServer.Procedural.Importing.Native.Using.Restore to perform an import.

NOTE: In the case of a full database restore the old database will need to be completely dropped before the backup can be restored. Partial restores do not require this.