- Newest
- Most votes
- Most comments
Hi there, there's a few ways you can do this and the following are some of the considerations factors:
- Upon job succeed, do you need to trigger any other workflows apart from the stored procedure
- Duration of stored procedure
- Frequency of the job running
Assuming you aren't triggering any other workflows and the duration + frequency of the jobs isn't high, you can stick to the lambda approach since it's serverless and can scale together with your glue job . Stored procedure wise, you can store them directly within the MySQL RDS DB by making some config changes to your DB - you can follow this guide here .
If your considerations are different and you might require to trigger additional workflows etc, high duration etc. You can consider the use of event bridge(enabling fan out architecture upon glue job completion) and other AWS services (i.e code build) for your solution.
Here is a technical guide that demonstrates how to use the combination of CodeBuild and EventBridge rules to schedule and run functions or stored procedures on an RDS for PostgreSQL database instance.
Hope it helps. Cheers.
Relevant content
- asked 2 years ago
- asked 8 months ago
- AWS OFFICIALUpdated 4 months ago
- AWS OFFICIALUpdated 2 years ago
- AWS OFFICIALUpdated 3 months ago
- AWS OFFICIALUpdated a year ago
Thanks a lot for your answer; here are some facts about my job :
Which scenario is better for these facts from the above? if lambda what would be the responsibility of lambda here? only connect to my rds DB and invoke the stored procedure?
if lambda? => if lambda invokes my stored procedure and may stored procedure take say 10 min to finish in this case lambda remain blocked
for 10 min until it gets a response back from the stored procedure ? or it invokes the stored procedure and stops?
Thanks a lot.