DEV Community: S.B.Shanjaiy

Day 1 AWS

S.B.Shanjaiy — Mon, 31 Mar 2025 09:26:29 +0000

Create a file in aws instance and push it to S3 bucket:

Open Putty and in connection->ssh->auth->credentials browse the location of pem.ppk file

copy the public ipv4 address and paste it in putty host name and click open

Three Tier Architecture Part-7

S.B.Shanjaiy — Mon, 31 Mar 2025 09:26:06 +0000

Three Tier Architecture Part-6

S.B.Shanjaiy — Mon, 31 Mar 2025 09:24:40 +0000

Three Tier Architecture Part-5

S.B.Shanjaiy — Mon, 31 Mar 2025 09:23:25 +0000

Three Tier Architecture Part-4

S.B.Shanjaiy — Mon, 31 Mar 2025 09:21:29 +0000

Three Tier Architecture Part-3 Database

S.B.Shanjaiy — Fri, 24 Jan 2025 10:09:49 +0000

Lets Start with creating subnet group for our database , open RDS and in that open subnet group

Next add the Availability zone and add the private subnet 1 and 2 which is in app tier

Lets create the database,first select standard create and select engine as aurora

Select template as dev/test

select self managed and click aurora standard

Select the Vpc and security group for the db

Now click create database

Three Tier Architecture Part-2 Vpc and Subnet

S.B.Shanjaiy — Fri, 24 Jan 2025 06:33:23 +0000

Welcome to Part-3 of Three tier Architecture.
First create a Vpc and give 10.0.0.0/16 in ipv4 and create the Vpc

Next create 4 private subnet and 2 public subnet for the aws-3-tier-workshop and create tag if needed as shown below

Now we have completed creating all 6 subnets

Next we have to create internet gateway as shown below

Next we are going to attach the Internet gateway to the vpv as shown below:

Lets now create a NAT gateway for both public subnets

Next we have to craete route table

Next we have to edit the route table by adding the internet gateway and save it as below

Now on subnet asssciation and add the 2 public subnet and save it:

Next we have to create 2 route table for 2 nat gateways

Now edit the route table and add the both nat gateway 1 and 2 for both the route table

Next in subnet association connect the first private subnet to first route table and the 2nd private subnet to the second route table

next we have to create 5 Security groups,first lets create security group for internet facing lb ,edit onlt the outbound rules

next lets create security group for web tier , edit only the inbound rules

Next create security group for internal load balancer and in inbound rules connect the security group of web tier

Next we have to create 4th security group for the app-tier we have to connect the security group of internal load balancer and in custom tcp we have to give port 4000

Next lets create the 5th security group connect to the db and the internal load balancer

Now we have completed the Part-3 of 3-Tier Architecture, See you in Part-4

Three Tier Architecture Part-1 S3 Bucket Creation And Iam Role

S.B.Shanjaiy — Fri, 24 Jan 2025 06:02:37 +0000

First use git clone or download the zip file from github(https://github.com/aws-samples/aws-three-tier-web-architecture-workshop.git)

Now create a S3 bucket to avoid errors us us-east-2 as aws account region with bucket versioning enabled

Next head to dashboard and open Iam to create a new Iam role for S3

Click next and add the following role:

AmazonSSMManagedInstanceCore
AmazonS3ReadOnlyAccess

Next create role name and click createrole

Now we have successfully created a role our S3

Amazon Elastic Inference

S.B.Shanjaiy — Thu, 23 Jan 2025 02:29:05 +0000

1.Service Overview:
Service Name: Amazon Elastic Inference
Logo:
Description: Amazon Elastic Inference allows you to attach just the right amount of GPU-powered acceleration to any Amazon EC2

2.Key Features:
Top Features:
Cost Efficiency
Scalability
Flexible GPU Allocation
3.Use Cases:
Real Life Applications:
Object detection
Facial recognition
Example:
Camera footage
4.Pricing Model:
The hourly rate applies only to the accelerator.

5.Comparison with similar service:
Choose Elastic Inference if you need cost-efficient, scalable GPU acceleration for inference on AWS where Google TPU offers high computational power.

6.Benefits and Challenges:
Advantages:Elastic Inference allows you to attach just the right amount of GPU power for inference,instead of paying for a full GPU instance when it's unnecessary.
Limitations:EI isn't portable to other platforms like Azure or Google Cloud.EI accelerators can only be attached to specific EC2 instance(e.g.,T3,M5,C5) and are not available forall other instance types.

7.Case Study:
Netflix attached Elastic Inference accelerators (such as eia2.medium and eia2.large) to their EC2 instances. This allowed them to effectively use the GPU parts of the recommendation system to the accelerators, while using cheaper CPU-based instances for less demanding tasks.