0

We have a Abinitio graph in On-prem, which relies on a configuration file to get few key inputs. To migrate this graph to AWS, we are looking for a better way to handle this configuration info.

This configuration info will be setup one time and it will get updated by other Abinitio process in runtime.

Below are more details:

Sample Configuration file:-

vi test.dat
Active_Flag|Y
Compress_File|N
Cut_Off_Date|20231231

Graph parameters relying on the Configuration File:-

COMPRESS_FLAG : $(cat test.dat| grep Compress_File | awk -F'|' '{print $2}'
CUT_OFF : $(cat test.dat| grep Cut_Off_Date | awk -F'|' '{print $2}'

We have seen few references in AWS wrt Named profiles. But not sure, if we can update the profile in runtime.

John Rotenstein
  • 241,921
  • 22
  • 380
  • 470
Prakash
  • 1
  • 2

1 Answers1

0

Named profiles usually refer to AWS CLI profiles configured locally for AWS CLI or client libraries to use to call AWS APIs. My understanding is that you want to store/update some configuration file, so your cloud-based application can access it.

Quick thought is to use Amazon Simple Storage Service (Amazon S3) or Amazon DynamoDB.

S3 can be literally used to store/update files. It also supports version history.

DynamoDB is a no-SQL database. If your usecase requires higher concurrency in terms of config file updates, your might leverage DynamoDB transactions to avoid potential conflicts.

charlesz
  • 291
  • 5