0

I have only found one answer that suggested tagging the tables. Since I am picking up an existing system and not building from scratch this would be labor intensive.

I have scoured AWS for a week and found no tangible steps for creating/maintaining UAT for DynamoDB.

The Java, local Dynamo does not persist data.

I would be grateful for any information regarding establishing a sustainable UAT environment for DynamoDB.

rkralston
  • 386
  • 1
  • 5
  • 20

1 Answers1

0

At my current employer we create a separate AWS environment for UAT. The separate environment is done by having another AWS account. This account can be used for UAT. So in our code, if the keys are for the UAT account then it would access the UAT DynamoDB databases. If the keys are for production, or development, then it would access the production or development DynamoDB databases.

This has some pros and cons. One con is that you have to manage multiple AWS accounts. But that's also a pro on that you can easily assess your AWS costs per account.

One pro is that you can keep your table names intact across environments. You do not need a "myTableNameUat" and "myTableNameDev"... its just "myTableName", and the keys which are passed to your application at deployment will determine which environment is used.

Jose Martinez
  • 11,452
  • 7
  • 53
  • 68
  • So far, these are the only two options I have seen. Do you have any tools for creating UAT data? – rkralston Jun 16 '15 at 19:33
  • Creating UAT data is a whole other major aspect of UAT. Three techniques I have seen for this are: create some UAT data by actually using the UAT components manually, generate UAT data via software that does this, borrow data from production (but make sure to clean it up so as not to cause any security or privacy issues). I have done all three of these methods throughout my career. – Jose Martinez Jun 17 '15 at 13:31
  • Yes, those are pretty much the highlevel options. However, I am asking a more detailed question. For example, to do this in SQL, I can back up a database and restore it to the UAT environment running a few scripts to scrub unsecured data. The process is rather simple given the backup/restore tools. My question here is to determine if a toolset exists for Dynamo to allow you to do something similar. My intuition is that nothing exists. The closest thing I have found is the tool for copying data to EC2 and then back. This seems like a kludge, but the most likely answer. Thanks – rkralston Jun 17 '15 at 22:41
  • You could use AWS datapipeline to do the backup / restore for you. You can scrub the data too using a hive activity. Read more about it here - http://aws.amazon.com/datapipeline/ – Rohit Kulshreshtha Jun 23 '15 at 07:23