Spot instanc3 for Availability?

[AWSPro21]

<div>Question asks how to ensure performance in case of AZ outage th3n recommends spot instances, seriously? Also the answer makes no mention of ASG but says other options are wrong because they don’t mention ASG.. 🤔 </div>

A digital media publishing company hired a solutions architect to manage its online portal, which is composed of an Application Load Balancer and Amazon EC2 instances deployed across multiple Availability Zones. The architecture uses a combination of Reserved EC2 Instances to handle the steady-state load and On-Demand EC2 Instances to handle the peak load. Currently, the web servers operate at 90% utilization during peak load.

<div>Which of the following is the most cost-effective option to enable the online portal to quickly recover in the event that one of the Availability Zones is unavailable during peak load?</div><div>
</div><div>To handle the peak load, launch an Auto Scaling group of Reserved instances on all Availability Zones.</div><div>To handle the peak load more effectively, use a combination of Spot and On-Demand instances on all Availability Zones.</div><div>To handle the peak load more effectively, use a combination of Reserved and On-Demand instances on all Availability Zones.</div><div>To handle the peak load in the most cost-effective manner, launch a Spot Fleet of EC2 instances with a diversified allocation strategy on all Availability Zones instead of On-Demand instances.</div><div>Incorrect</div><div>The focus of the scenario is how to properly handle the peak load of your application in the most cost-effective manner since the steady-state load is already handled by the Reserved EC2 instances. Spot Instances provide the cheapest compute capacity to handle the peak load.</div><div>
</div><div>
</div><div>
</div><div>To avoid interruption to your Spot instances, you can actually set up a diversified spot fleet allocation strategy in which you are using a range of different EC2 instance types such as c3.2xlarge, m3.xlarge, r3.xlarge et cetera instead of just one type. This will effectively increase the chances of providing a more stable compute capacity to your application. Therefore, in the event that there is a Spot interruption due to the high demand for a specific instance type, say c3.2xlarge, your application could still scale using another instance type such as m3.xlarge or r3.xlarge.</div><div>
</div><div>Hence, the correct answer in this scenario is the option that says: To handle the peak load in the most cost-effective manner, launch a Spot Fleet of EC2 instances with a diversified allocation strategy on all Availability Zones instead of On-Demand instances.</div><div>
</div><div>The following options are incorrect because the scenario requires a highly available and cost-effective architecture to allow the application to recover quickly, hence, using an Auto Scaling group is a must to handle the peak load and improve both the availability and scalability of the application:</div>

[Carlo-TutorialsDojo]

Hello AWSPro21,

Thanks for the feedback.

I understand how the given explanation can be confusing. To clarify, an ELB can work with a Spot Fleet even without an ASG. Both ASG and Spot Fleet use AWS Auto Scaling, where you can specify a minimum capacity to be maintained and a maximum capacity to scale up to. A key difference between the two is that AWS pretty much manages the instance lifecycle in a Spot Fleet. In contrast, with an ASG, you have some control over the lifecycle through lifecycle hooks.

Just FYI, AWS recommends using EC2 Fleet instead of Spot Fleet. These two are practically similar in core functionalities, with Spot Fleet relying on legacy APIs. Nonetheless, we continue to use the term ‘Spot Fleet’ since it appears more frequently than ‘EC2 Fleet’ in the SA Pro certification exams.

With all that said, we acknowledge there are some areas needing improvement, which we will address.

Let me know if you have any further clarifications.

Regards,

Carlo @ Tutorials Dojo