You are designing a deployment technique for your new applications on Google Cloud. As part of your deployment planning, you want to use live traffic to gather performance metrics for both new and existing applications. You need to test against the full production load prior to launch. What should you do?
To gather performance metrics for both new and existing applications and test against the full production load prior to launch, you should use A/B testing with traffic mirroring. This approach allows you to mirror live traffic to the new application version while still serving it to the existing version. This ensures that both versions are subjected to the same real-world conditions, providing a true comparison of performance under full production load.
D due to "full production load"
"You need to test against the full production load prior to launch" It's impossible with canary. "A/B testing with traffic mirroring during deployment" is the only one possibility we have to test the entire traffic before the roll out.
you want to use live traffic to gather performance metrics for both new and existing applications
Canary will only redirect a small portion of the traffic, while A/B with mirroring will test the new version in full load
A/B testing with traffic mirroring during deployment. This technique allows you to divert a portion of the live production traffic to the new application version while still serving the majority of the traffic to the existing version. By comparing the performance metrics of both versions under real-world conditions, you can assess the impact of the new deployment on your application’s performance and stability.
D is correct
Correct Answer is Shadow test pattern, as it is not in the option selected D https://cloud.google.com/architecture/application-deployment-and-testing-strategies#shadow_test_pattern
D is my answer
Canary can test live production traffic on production (Answer A) https://cloud.google.com/architecture/application-deployment-and-testing-strategies#key_benefits_4
is it possible that right answer is not present. To me the answer is "Shadow Test Pattern" https://cloud.google.com/architecture/application-deployment-and-testing-strategies#shadow_test_pattern. Do you agree ?
Correct answer is A: Canary deployment is a technique to reduce the risk of introducing a software update in production by slowly rolling out the change to a small subset of users before making it available to everybody. This deployment technique is one where the SRE of an application development team relies on a router or load balancer to target individual routes. They target a small fragment of the overall user base with the newer version of the application. Once this new set of users are have used the application important metrics will be collected and analyzed to decide whether the new update is good for a full scale rolled to all the users or whether it needs to be rolled back for further troubleshooting.
A/B Testing with Traffic Mirroring: This approach allows you to send a portion of your live production traffic to both the new and existing versions of your application simultaneously. This provides a realistic test environment with real-world load, allowing you to gather performance metrics for both versions side-by-side. Traffic mirroring ensures that the new version is tested under the same conditions as the existing version, providing a more accurate comparison.
A. Canary Deployment: While canary deployment is useful for gradual rollouts and testing, it primarily focuses on testing the new version in a limited environment. It doesn't necessarily provide a full production load test. B. Blue/Green Deployment: Blue/green deployment involves switching traffic between two identical environments. While it allows for zero downtime deployments, it doesn't provide a way to test both versions simultaneously with live traffic. C. Rolling Updates Deployment: Rolling updates gradually replace instances with new versions. This approach doesn't offer a way to test both versions with live traffic simultaneously.
D is the answer.
D, the question requires more than just "a load" rather the "full load" the only strategy where this happens is A/B testing
After giving a deliberate thought, I think it's option D. The keyword here is 'gather performance metrics,' and as everyone must know A/B testing's whole purpose is to gather performance metrics.
Vote A
Full production - A/B testing, option D
Agree with Option A