Domain 4: Design Cost-Optimized Architectures Flashcards
You work for an automotive company that has a small estate on AWS, but the majority of their assets are hosted in-house at their own data center. They are now looking to save money by moving more and more real estate to AWS and have started creating multiple AWS accounts in the same Region. As part of their expansion strategy, they plan to deploy multiple VPCs across these AWS accounts. They currently have one Direct Connect connection installed between their on-premises data center and AWS. Now that they have multiple production accounts and a growing number of VPCs, they will need to connect these to the on-premises data center using a dedicated connection. What is the most cost-effective way of doing this?
Use a VPN concentrator to connect the AWS accounts back to the on-premises data center.
Provision an AWS VPN CloudHub and connect the AWS accounts directly back to the Direct Connect connection via a VPN connection.
Create a new Direct Connect gateway and set this up with the existing Direct Connect connection. Set up a transit gateway between the AWS accounts and connect the transit gateway to the Direct Connect gateway.
Provision a new Direct Connect connection for each AWS account and connect it back to your on-premises data center.
Provision an AWS VPN CloudHub and connect the AWS accounts directly back to the Direct Connect connection via a VPN connection.
If you have multiple AWS Site-to-Site VPN connections, you can provide secure communication between sites using the AWS VPN CloudHub. However, this is not required in this scenario, since you can take advantage of the existing Direct Connect connection and associate it to an AWS Direct Connect gateway with a transit gateway to connect multiple VPCs in the same Region.
Selected
Create a new Direct Connect gateway and set this up with the existing Direct Connect connection. Set up a transit gateway between the AWS accounts and connect the transit gateway to the Direct Connect gateway.
You can associate an AWS Direct Connect gateway with a transit gateway when you need to connect multiple VPCs in the same Region. AWS Documentation: Direct Connect gateways.
You run a popular retro gaming merchandise retail platform on AWS. Over the past year and a half, you’ve noticed that your traffic has distinct daily and weekly patterns. For example, there’s a surge in traffic during weekday business hours and a huge drop during the weekends. To add to the issues, your application takes a considerable amount of time to initialize, causing a noticeable latency impact during scale-out events. You need to ensure that your infrastructure scales in anticipation of these patterns. Which AWS Auto Scaling feature would best address this scenario?
Manual scaling
Dynamic scaling
Predictive scaling
Scheduled scaling
Predictive scaling
Predictive scaling uses machine learning to forecast traffic and capacity needs. Predictive scaling is more flexible and can adapt to changes in traffic patterns, which is why it is the best choice for the given scenario. If your traffic spikes during weekdays and drops during weekends, predictive scaling can automatically scale your resources in anticipation of these patterns. This will help to ensure that you have enough resources to handle peak traffic without causing latency issues.
Scheduled scaling
Scheduled scaling allows you to increase or decrease the number of instances in your Auto Scaling group based on a specific schedule. While it can be set up to scale based on known traffic patterns, it doesn’t automatically adjust to changes in those patterns. In the given scenario, while scheduled scaling can handle the predictable surge in traffic during weekdays and drop during weekends, it would not be as adaptive as predictive scaling.
Selected
You have developed an AI-powered app that is used to predict the prices of cryptocurrency in real time. The app requires low latency and high throughput storage performance for processing training sets. You need to archive the completed processed training sets on storage that is as cost effective as possible, but can still maintain immediate access. What two storage solutions should you use? CHOOSE 2
Amazon FSx for Lustre for processing training sets
Amazon Elastic File System for archiving completed processed training sets
AWS Storage Gateway for processing training sets
Amazon S3 Glacier Instant Retrieval for archiving completed processed training sets
Amazon FSx for Lustre for processing training sets
This is the best performing option for processing training sets in this scenario.
AWS Storage Gateway for processing training sets
This is not a technically viable option for processing training sets.
Selected
Amazon S3 Glacier Instant Retrieval for archiving completed processed training sets
Amazon S3 Glacier Instant Retrieval is the lowest-cost storage for long-lived data that is rarely accessed and requires retrieval in milliseconds. AWS Documentation: Amazon S3 Glacier Instant Retrieval.
Selected
You work for a private library that is digitizing its collection of ancient books. The library wants to store scans of each book in the cloud at the cheapest rate possible. The files will be accessed only occasionally, but will need to be retrieved instantly. What is the most cost-effective way to achieve this?
S3 Infrequent Access
Elastic Block Storage (EBS)
Elastic File System (EFS)
S3 Standard
S3 Infrequent Access
S3 Infrequent Access is suitable for files that will be accessed only occasionally but require instant retrieval.
You have a steady application serving around 3,000 customers that needs to be migrated to AWS. Based on historical data, traffic and usage has not grown very much in the past 24 months and you expect the application to remain steady for the next 3 years. You need to run the application on EC2. What is the most cost-effective EC2 instance type to use?
Dedicated Instances
On-Demand Instances
Reserved Instances
Spot Instances
Reserved Instances
Using Reserved Instances is the most cost-effective solution for running EC2 instances over long periods of time.
You are working for a small startup that wants to design a content management system (CMS). The company wants to architect the CMS so that the company only incurs a charge when someone tries to access their content. Which services could be used to help create a CMS while prioritizing pay-per-use pricing services as to allow for minimal cost? CHOOSE 3
EC2
Application Load Balancer
DynamoDB
API Gateway
S3
DynamoDB
DynamoDB is a NoSQL database service that provides fast and predictable performance with seamless scalability. It operates on a pay-per-read and pay-per-write basis, which is ideal for a CMS that desires to minimize costs when the system is not in use. By storing metadata or content within DynamoDB, the startup only incurs charges when read/writes occur.
Selected
API Gateway
API Gateway is a fully managed service that makes it easy for developers to create, publish, maintain, monitor, and secure APIs at any scale. It operates on a pay-per-use model, charging for the number of API calls made, making it a cost-effective solution for a CMS architecture where costs are incurred only when content is accessed. API Gateway’s integration with other AWS services like Lambda can help in building a serverless CMS, further aligning with the pay-per-use pricing model.
Selected
S3
Amazon S3 is an optimal choice for storing and retrieving any amount of data, making it a suitable backbone for a CMS. The pay-as-you-go pricing model of S3 aligns well with the requirement of incurring charges only when content is accessed.
Selected
You work for a small startup that has a shoestring budget. You accidentally leave a large EC2 instance running over a few days and are hit with a huge bill. You need to prevent this from happening in the future. What should you do?
Create a billing alarm to monitor your AWS charges for when they go above a certain threshold.
Use AWS Trusted Advisor to notify you whenever an EC2 instance has been running for more than 24 hours.
Enable CloudFormation to alert you when any EC2 instance has been running for more than 24 hours.
Enable AWS CloudTrail to terminate any EC2 instance that has been running for more than 24 hours.
Create a billing alarm to monitor your AWS charges for when they go above a certain threshold.
This would be your best course of action.
Selected
You work for a large advertising company that is moving its videos and photos to AWS. The size of the migration is 70 terabytes, and it needs to be completed as quickly and cost-effectively as possible. What is the best way to achieve this?
AWS File Gateway
An AWS Snowball Edge Storage Optimized device
AWS Direct Connect
AWS Storage Gateway
An AWS Snowball Edge Storage Optimized device
This would be the cheapest way to transfer large amounts of data.
You work for an insurance company that has just been merged with two other insurance companies. All companies have production workloads on AWS using multiple AWS accounts. Which of the following is something you could recommend to your boss to immediately start saving money?
Migrate all AWS accounts to a single AWS account and close the migrated accounts.
Run Amazon Macie to identify where you can save costs.
Use AWS CloudTrail to start keeping track of what you are spending.
Create a root AWS account using AWS Organizations and connect all subsequent AWS accounts to the Organization. You can then take advantage of consolidated billing.
Create a root AWS account using AWS Organizations and connect all subsequent AWS accounts to the Organization. You can then take advantage of consolidated billing.
Using consolidated billing, you can pool your AWS resources to lower your total costs.
You work for a popular streaming service that runs its NoSQL backend in-house on large Cassandra clusters. You recently had a major outage and realize you need to migrate your Cassandra workload on to something more reliable, such as the AWS Cloud. You do a cost analysis and realize that, in the long run, this will probably save the company a lot of overhead fees. You need to select a Cassandra-compatible service on which to run your workloads. Which service should you select?
Amazon Keyspaces
Amazon DocumentDB
Amazon Keystone
Neptune
Amazon Keyspaces
This is a Cassandra-compatible database and is the best choice for this scenario.
You host a web application on Amazon EC2 that contains a large number of files that are infrequently accessed. Currently, the files are hosted on provisioned IOPS; however, due to budget cuts, your manager asks you to move the files to a more cost-effective solution. What storage solution should you choose?
Use a Cold HDD (sc1).
Use an Elastic Block Storage General Purpose SSD (gp3).
Use an S3 Infrequent Access storage bucket. Create a role in IAM granting S3 access and attach this role to your EC2 instance.
Use a Throughput Optimized HDD (st1).
Use an S3 Infrequent Access storage bucket. Create a role in IAM granting S3 access and attach this role to your EC2 instance.
This would be the cheapest way to store your data in this scenario.
A Fintech startup has a small application that receives intermittent and random traffic. At some points, it may not receive any traffic at all; at other times, it might receive tens of thousand of queries at once. You need to rearchitect the application for the AWS cloud using a relational database. What database technology would best suit your needs while keeping costs at a minimum?
DynamoDB
NeptuneDB
Aurora Serverless
RDS for MySQL
Aurora Serverless
This is the best answer as it keeps cost low while being a relational database.
What is the most cost-effective architecture for a front-facing website, assuming a peak load of 500 users per hour will be accessing the site?
A fleet of EC2 instances behind a Network Load Balancer connected to an RDS instance with multiple read nodes
An Elastic Beanstalk configuration using Auto Scaling and EC2
A serverless website using API Gateway, Lambda, and DynamoDB
An Elastic Kubernetes Service cluster
A serverless website using API Gateway, Lambda, and DynamoDB
Given this short scenario, this would be the most cost-efficient and scalable solution.
