Scribd
Upload
26 views3 pages

Discussion 1 Cloud Computing Technologies

The document discusses selecting AWS services and resources for building scalable web applications and machine learning workloads. It recommends choosing EC2 instance types and sizes that provide scalability and cost-effectiveness, using a mix of AWS storage services, and integrating additional services to improve performance, redundancy, and scalability.

Uploaded by

hassanzamir702
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
26 views3 pages

Discussion 1 Cloud Computing Technologies

The document discusses selecting AWS services and resources for building scalable web applications and machine learning workloads. It recommends choosing EC2 instance types and sizes that provide scalability and cost-effectiveness, using a mix of AWS storage services, and integrating additional services to improve performance, redundancy, and scalability.

Uploaded by

hassanzamir702
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Discussion 1

Student Name:

Student ID:

Date:

Professor Name:
Scenario 1
When it comes to building a scalable web application, our main priority should be selecting the
most suitable EC2 instance types, storage solutions, and AWS services to guarantee optimal
performance, redundancy, and scalability.

EC2 Instance Type and Size Selection:

 Criteria: Our goal is to find instances that can effectively handle a surge in user traffic
without compromising performance and cost-effectiveness.
 Decision: Choose EC2 instances with horizontal scalability and auto-scaling features, like
the AWS EC2 T3 or M5 instance families. These instances provide a well-rounded
combination of compute power, memory, and network performance, making them ideal
for managing a range of workloads.
 Size: Ensure that the chosen instance sizes can effectively handle peak traffic without
incurring unnecessary costs during low-traffic periods.

Storage Solutions:

 Criteria: The storage solution must be durable, scalable, and high-performing to handle
data storage and management for the application effectively.
 Decision: Opt for a mix of AWS storage services to cater to various storage needs:
o Amazon RDS (Relational Database Service) is a powerful tool for managing
relational databases, providing data consistency and scalability for transactional
data.
o Amazon S3 is a great option for storing static assets such as images, videos, and
documents. It provides high durability, scalability, and low-latency access.
o Utilize Amazon ElastiCache to efficiently cache frequently accessed data,
enhancing application performance and decreasing strain on the database.
o Amazon EBS is a reliable and versatile solution offering block-level storage
volumes for EC2 instances. It is designed to store application data and system files
securely.

Integration of AWS Services:

 Objective: Integrate AWS services to enhance the web application's performance,


redundancy, and scalability.
 Recommendation: It is advisable to incorporate the following AWS services to improve
application performance and reliability:
o Utilize Amazon Route S3 for DNS web service to enhance availability and fault
tolerance through traffic routing, load balancing, and health monitoring.
o Utilizing Amazon CloudFront as a content delivery network (CDN) enables the
caching and delivery of content at edge locations across the globe, resulting in
reduced latency and an enhanced user experience.
o Consider using Amazon SQS or Amazon SNS to improve your application's
scalability and fault tolerance by enabling decoupling and asynchronous
communication between its components.
o Utilizing AWS Lambda allows for serverless computing, where code can be
executed in response to events or triggers. This helps to minimize operational
overhead and enables auto-scaling based on workload demands.

With careful selection of EC2 instance types, proper storage solutions, and effective integration
of AWS services, we can create a highly scalable web application that performs optimally and
provides redundancy on the AWS cloud platform.

Scenario 2
Let's start our machine learning or deep learning project on AWS by carefully choosing the most
appropriate EC2 instance types and storage solutions.

EC2 Instance Types:


 Considering the varied nature of ML/DL workloads, it is crucial to consider both flexibility
and raw computational capability. Choosing a flexible instance type, such as the AWS
EC2 M5 family, can offer a well-rounded strategy.
 The M5 instances provide a blend of computational, memory, and networking resources
that are well-suited for a diverse array of machine learning and deep learning activities.
They are highly suitable for applications involving inference workloads, model servicing,
and lightweight training.

Storage options:
 Amazon S3 is our preferred option for storing datasets and model artifacts because of its
high durability, scalability, and cost-effectiveness.
 When it comes to training, where achieving high input/output throughput is essential,
Amazon Elastic Block Store (EBS) volumes with provided IOPS (io1) can deliver the
required performance enhancement for temporary storage requirements.
 Although our primary focus is on inference and lightweight training, it may still be
advantageous to leverage Amazon EFS for shared file storage across instances due to
its simplicity and scalability.

Optimization and Cost Considerations:


 Although Spot Instances and Auto Scaling are potent tools, they may not be critical for our
workload, primarily emphasizing inference and lightweight training rather than extensive
batch training.
 Nevertheless, monitoring resource use and cost using Amazon CloudWatch to guarantee
effective resource allocation and cost control is crucial.

To create a flexible cloud-based environment for our machine learning or deep learning
project on AWS, we can choose versatile EC2 instances from the M5 family, use Amazon S3
for long-term storage, Amazon EBS for temporary storage, and potentially incorporate
Amazon EFS for shared file storage.

You might also like