site-logo

JAWS PANKRATION 2024

site-logo
HomeNewsTimetableCfPCommitteePromotionFollow UpPrivacy Policy

Cloudy with a chance of Precision: Deploying Scalable Weather forecasting apps on AWS EKS

Lv300

Lv300

8/24/2024 09:00 (UTC)
Session Info

Weather forecasting demands scalability, precision and reliability.

This session will dive deep on how we can build such apps using Python, create images using Docker/Podman and scale it using AWS EKS service.

We will discuss strategies for integrating diverse meteorological data sources into AWS ecosystem for data consistency and accuracy.

Join us to discover best practices in cloud native application development, insights into integrating real time weather data APIs & strategies for ensuring high availability and performance in weather prediction services.

Neetu Mallan

Neetu Mallan

- AWS Community Builders -

- AWS User Community Leaders -

Session Category
Computing
Container
AWS Services
AWS EKS
AWS Codebuild
AWS S3
RDS
Session Materials
    Session Archive
    Session Summary (by Amazon Bedrock)
      The speaker discusses deploying a weather forecasting app on AWS EKS (Elastic Kubernetes Service) using containerization. They start by introducing a Python-based weather app that uses the OpenWeatherMap API. The app allows users to select a city, temperature unit, and graph type to display weather information. The speaker then explains how they containerized the application using Docker, following cloud-native best practices such as multi-stage builds. This approach reduced the image size from 900MB to 587MB. Next, they discuss the AWS services used in the deployment: 1. ECR (Elastic Container Registry) for storing container images 2. Secrets Manager for securely storing the API key 3. EKS for container orchestration 4. Load Balancer for accessing the application The application architecture is described, showing how the dockerized app is uploaded to ECR, orchestrated in EKS, and accessed via a load balancer. The Secret Store CSI driver is used to retrieve the API key from Secrets Manager. The speaker then demonstrates the deployment process, showing: 1. Creating a manifest file for multi-architecture images (AMD64 and ARM64) 2. Deploying the application using Kubernetes manifests 3. Mounting the API key as a Kubernetes secret 4. Accessing the deployed application through a Windows RDP session The demo concludes by showcasing the running weather app, displaying temperature, wind speed, humidity, and other weather information for a selected city. In conclusion, the speaker emphasizes how containerization and EKS can be leveraged to deploy and run applications efficiently in AWS.

    Back

    ©JAWS-UG (AWS User Group - Japan). All rights reserved.