Skip to content

Home

Jump to bottom
aksrajvanshi edited this page Mar 20, 2020 · 16 revisions

Welcome to the pika-pika wiki!

Please find the architecture diagram at https://drive.google.com/file/d/1qHBmQgGoLnbUws8xLXEn1bedcwGb8UPz/view?usp=sharing

Login into https://iu.jetstream-cloud.org/auth/login/ and ensure your allocation is TG-CCR*****43. Make sure it ends with the number 43 and download OpenRC File v3 from the top right corner dropdown.

Ensure the Docker is running on your system. Firstly run the following command,

docker pull akshayrajvanshi1992/pika-pika-ubuntu:latest

Then, after that run the following command:

docker run --name pika-ubuntu -v <folder-path-for-volume>:/data -it akshayrajvanshi1992/pika-pika-ubuntu "/bin/bash"

The "folder-path-for-volume" is the folder which your container accesses. It would be better if you make a folder on your Desktop for easy access.

When you have to login later, run this command to enter inside your container you just created:

docker container start pika-ubuntu
docker exec -it pika-ubuntu /bin/bash

Copy your OpenRC v3 file into the volume folder you created for pika-ubuntu container. Then log in again into the docker container with the exec command given above. Then run the following command.

cp -p /data/<OpenRC file> ~/openstack/

Dockerized Kafka

Firstly, we add a dockerized version of Kafka and Zookeeper to the project.

Usage

For macOS, we run a simple command.

make

For Windows, we have a simple command here to run from the command prompt or you can just double click the windows-docker-makefile.bat file.

windows-docker-makefile.bat

After the docker has built the container, login into the container with the following command. But first, you will have to know the container ID. You can perform the following commands.

docker ps -a
docker exec -it <Container ID> /bin/bash

After you are inside the docker container bash, perform the following commands:

kafka-topics --zookeeper 127.0.0.1:2181 --create --topic data-retrieval-service --partitions 1 --replication-factor 1
kafka-topics --zookeeper 127.0.0.1:2181 --create --topic model-execution-service --partitions 1 --replication-factor 1
kafka-topics --zookeeper 127.0.0.1:2181 --create --topic user-management-service --partitions 1 --replication-factor 1
kafka-topics --zookeeper 127.0.0.1:2181 --create --topic post-processing-service --partitions 1 --replication-factor 1
kafka-topics --zookeeper 127.0.0.1:2181 --create --topic session-management-service --partitions 1 --replication-factor 1
kafka-topics --zookeeper 127.0.0.1:2181 --create --topic api-gateway-service --partitions 1 --replication-factor 1
kafka-topics --zookeeper 127.0.0.1:2181 --create --topic front-end-data-retrieval-service --partitions 1 --replication-factor 1

Data Retrieval Service

This microservice takes into account location, date of the query being made and gets NEXRAD radar and version data of the file we need to predict weather of the given location.

Installation

Ensure that you have maven installed in the system. If you don't have maven, download it from the following link:

https://maven.apache.org/download.cgi

Usage

cd DataLinkRetrieve/
mvn clean install && mvn spring-boot:run

Ensure that the port 8080 is not busy.

For the following Microservices

API Gateway Microservice, Frontend-data-retrieval-microservice, Post-processing-microservice, Session-management-microservice, User-management-microservice

Installation

Ensure that you have npm installed in the system. If you don't have npm, download it from the following link:

https://www.npmjs.com/get-npm

Usage

Install the api gateway microservice by running the following command in the folder you cloned the microservice in using command-line :

cd <microservice-folder-name> (i.e. user-management-microservice)
npm install
node .

Model-Execution-Microservice

Installation

Install the model execution microservice by navigating to the following folder using command-line :

cd model-execution-microservice/model-execution-microservice/modelenv/Scripts

Run the command :

activate

Navigate two folders back using the following command :

../../pip install -r requirements.txt

Execute the python script using the following command :

python consumer.py

Frontend

Installation

In the project directory, you can run:

https://www.npmjs.com/get-npm

and then do

npm start
Clone this wiki locally