EV Charging Network - MongoDB NoSQL Extension

MongoDB implementation of an EV charging network analytics workload, migrated from an Oracle relational schema and documented for a Database + Models NoSQL assignment.

Project Context

Item	Details
Course	Database + Models, Semester 2
Faculty	ELTE Faculty of Informatics
Date	April 2026
Source DBMS	Oracle Database
Target DBMS	MongoDB Community Server
MongoDB database	ev_charging_nosql
Primary tools	Python, PyMongo, MongoDB Compass, mongosh

Overview

This repository contains the NoSQL extension of an EV Charging Network database project. It converts operational data from the Oracle source schema into MongoDB collections, then evaluates the resulting document model through index creation, document updates, and analytical aggregation pipelines.

The full submitted report is available as docs/EV_Charging_NoSQL_Project_Documentation.docx. It contains the migration rationale, screenshots, query outputs, Oracle vs MongoDB comparison, references, code listings, and language-model usage disclosure.

What This Repository Demonstrates

• Oracle-to-NDJSON export using Python.
• MongoDB bulk import using PyMongo.
• A nine-collection MongoDB model for EV charging operations.
• Ten MongoDB operations covering DDL, DML, simple aggregation, and complex aggregation.
• Evidence-backed comparison between the relational Oracle design and the MongoDB implementation.

Data Model

The MongoDB database keeps the main Oracle entities as separate collections so that the migration remains traceable to the original relational design while still supporting document-oriented query patterns.

Collection	Documents
charging_sessions	100,000
meter_readings	110,000
customers	10,000
vehicles	12,000
maintenance_tickets	3,000
stations	120
connectors	360
tariffs	12
session_audit_log	2

Generated NDJSON files and local execution logs are not committed because they are large or reproducible from the scripts.

Repository Structure

.
|-- scripts/
|   |-- export_oracle_to_json.py
|   `-- import_to_mongo.py
|-- queries/
|   |-- 00_run_all_queries.js
|   |-- 01_ddl_dml.js
|   |-- 02_simple_queries.js
|   `-- 03_complex_queries.js
|-- docs/
|   |-- EV_Charging_NoSQL_Project_Documentation.docx
|   `-- figures/
|-- EV Charging NoSQL Relationship Model.png
`-- README.md

Prerequisites

• Python 3.9 or newer.
• Oracle client access to the source database.
• MongoDB Community Server running locally on port 27017.
• MongoDB Compass or mongosh for query execution.
• Python packages:

pip install oracledb pymongo

Reproducibility Workflow

1. Export Oracle Tables

python scripts/export_oracle_to_json.py

The exporter writes one NDJSON file per source table into data/. Oracle credentials can be supplied with environment variables or entered interactively.

Variable	Default	Purpose
ORACLE_USER	prompted	Oracle username
ORACLE_PASSWORD	prompted	Oracle password
ORACLE_HOST	localhost	Oracle host
ORACLE_PORT	2001	Oracle port
ORACLE_SERVICE	aramis	Oracle service name

2. Import Into MongoDB

python scripts/import_to_mongo.py

Variable	Default	Purpose
MONGO_URI	mongodb://localhost:27017	MongoDB connection string
MONGO_DB	ev_charging_nosql	Target database name

3. Run The MongoDB Workload

Run the complete workload with mongosh:

mongosh mongodb://localhost:27017 queries/00_run_all_queries.js

Alternatively, open MongoDB Compass, connect to mongodb://localhost:27017, select ev_charging_nosql, and load the query scripts from the embedded shell.

Query Workload

#	Category	Operation
Q1	DDL	Create a compound index on charging_sessions(session_status, session_start)
Q2	DDL	Create a compound index on meter_readings(session_id, reading_time)
Q3	DML	Update one charging session by _id
Q4	DML	Run a conditional bulk update plus reversible insert/delete verification
Q5	Aggregation	Count sessions by status
Q6	Aggregation	Rank stations by connector count
Q7	Complex aggregation	Calculate monthly completed-session energy totals
Q8	Complex aggregation	Rank connectors by usage and revenue
Q9	Complex aggregation	Enrich unpaid sessions with customer details
Q10	Complex aggregation	Calculate average energy and cost by tariff

The query scripts are split by purpose, while queries/00_run_all_queries.js provides a single entry point for running the full sequence.

Documentation And Evidence

• docs/EV_Charging_NoSQL_Project_Documentation.docx is the current submitted report.
• docs/figures/ contains the evidence figures used in the report.
• queries/ contains the reproducible MongoDB workload.
• scripts/ contains the migration scripts referenced by the report.

Data And Security Notes

• Exported data files under data/ are intentionally ignored by Git.
• Local logs and regenerated artifacts should be treated as reproducible outputs, not source files.
• Database credentials are not stored in the repository.

Related Repository

The Oracle source project is maintained separately:

ev-charging-network-oracle-project - EV Charging Network Management System, Oracle Database implementation.

This MongoDB repository references the Oracle schema entities by name but does not import code from the Oracle repository directly.