A company needs to run a critical application on AWS. The company needs to use Amazon EC2 for the application’s database. The database must be highly available and must fail over automatically if a disruptive event occurs.
Which solution will meet these requirements?
To ensure high availability and automatic failover for the application's database on Amazon EC2, the best solution is to launch two EC2 instances in different Availability Zones within the same AWS Region, install the database on both instances, configure them as a cluster, and set up database replication. This configuration ensures that if one instance or one Availability Zone fails, the database can automatically fail over to the other instance in the different Availability Zone, maintaining high availability and minimal downtime.
Changing my vote to A. After reviewing a Udemy course of SAA-C03, it seems that A (multi-AZ and Clusters) is sufficient for HA.
what number of class ?
Here AWS defines HA, and uses the word cluster - AWS has several methods for achieving HA through both approaches, such as through a scalable, load balanced cluster or assuming an active–standby pair. - https://docs.aws.amazon.com/whitepapers/latest/real-time-communication-on-aws/high-availability-and-scalability-on-aws.html
The question states that it is a critical app and it has to be HA. A could be the answer, but it's in the same AZ, so if the entire region fails, it doesn't cater for the HA requirement. However, the likelihood of a failure in two different regions at the same time is 0. Therefore, to me it seems that C is the better option to cater for HA requirement. In addition, C does state like A that the DB app is installed on an EC2 instance.
but for C you need communication between the two VPC, which increase the complexity. With a should be enough for HA
The question doesn't ask which option is the most HA. It asks what meets the requirements.
Design for region failure? may as well design for AWS failure and put replica in GCP and Azure :v
And on-prem in multiple DCs and one in mars too :D
yep lol, even in the other questions, for HA you can use Multi-AZ
this is reading comprehension exam not a practical exam.
Option C proposes launching two EC2 instances in different AWS Regions and setting up database replication, with failover to a second Region. While this solution does provide geographic redundancy, it may introduce higher latency due to cross-region communication and data replication. Additionally, failover to a different Region typically involves more complex configurations and longer recovery times compared to failover within the same Region. While Option C may offer a level of redundancy, it might not provide the same level of responsiveness and automatic failover capabilities as Option A, which leverages Availability Zones within the same Region. In scenarios where low latency and rapid failover are critical, Option A is often preferred. However, if geographic redundancy is a top priority and the potential trade-offs in latency and failover time are acceptable, Option C could still be a viable solution.
A. High Availability - multiple Zones. Disaster Recovery - multiple Regions.
Option A suggests deploying two EC2 instances, each in a different Availability Zone within the same AWS Region. This ensures high availability by distributing the instances across multiple physically isolated locations. By installing the database on both EC2 instances and configuring them as a cluster, you create a highly available database setup where one instance can seamlessly take over if the other instance fails. Additionally, setting up database replication between the instances ensures data consistency and redundancy. If one instance fails, the other instance can continue serving requests without interruption.
Answer is C, because question mentioned about disruptive event occurs. when the whole region failed, it can not cover the scenario for HA
The term "disruptive event" implies it requires DR, HA is not sufficient.
C "if a disruptive event occurs"
A provides HA with 2 EC2 in two AZ with database replication
C for sure
What does "Configure the EC2 instances as a cluster" mean? The only "EC2 cluster" that I am aware of is a "cluster placement group". If that's the case, then all EC2 instances in that cluster must be in the same AZ. So option A would be invalid then.
Option B uses an AMI for backup and CloudFormation for automation, but doesn't provide high availability or automatic failover. Option C launches instances in different regions, which may not be necessary and may increase costs. Option D uses EC2 automatic recovery, which can recover an instance, but doesn't provide high availability or automatic failover.
Explanation: Option C provides a solution that ensures high availability by deploying EC2 instances in different AWS Regions. By setting up database replication and failing over the database to a second Region, you ensure automatic failover if a disruptive event occurs in one Region. Options A and B focus on high availability within a single AWS Region but don't address automatic failover to a different Region in case of a disruptive event. Option D uses EC2 automatic recovery, but it doesn't provide a solution for automatic failover to a different Region, which is necessary for ensuring high availability in case of a Region-wide failure.
C - I am a littel be wonder reading same explanation, becouse exist a vary big differance beetwen instance cluster and dababase cluster.
A is for HA. D is for DR
Perfect Answer is A
A is more correct to support both HA and Failover. C is only for Failover, not HA during the traffic.
C will be answer As it is a critical application