User

Signup

Login

Auth Router

Use JWT Token

Store Instance in Cache

Auth Login Router

Privilege Base

MongoDB

Requests (Get/Post)

Admin

User

Current

Analysis

CRUD

Approval

Service Provider

Service

Chat

School Food List

MongoDB

Output

Nokia

Ericson

Cisco

GPFS (file store location)

Standardization of format

CDP

Hive

Iceberg

Impala

Sahyog

Ozone S3 compliant Object Store Can be queried via Impala As of now, nothing in Ozone

35% data in Hive, 65% in Iceberg ~2X compression

This data volume is a challenge. Requirement is to store data and process data in efficient storage with higher compression rate and better query response time.

• Sources generate files in different formats. • A component to standardize the format and move to CDP via Kafka

• ETL using Pyspark • Output Parquet file to Hive

Front End for User Interaction Has Dashboards etc.