Data Modelling with

MongoDB
5th grade / V. Jahrgang

Quellen: linkedin
Aspects of Data Modelling
• Describing Entities
• Describing Relationsships
• Validatation of Data
• „Translation“ of these ideas into a logical concept to fullfill these
aspects (f.ex. Relational datamodel)

Status : 14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 2

noSQL-View on Data Modelling
• Describing Entities by describing objects and defining their properties
and the methods to change them
• Describing Relationsships
• Validata Data
• „Translation“ of these ideas into a logical concept to fullfill these
aspects , f.ex
• Json or xml
• Graphs and distances

Status : 14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 3

Logical concepts of Document based
Databases
• Collections
• Documents
• Subdocuments/embedded documents
• Arrays
• Document centric view = document is the single source of truth
• No/little normalization
• Joins are rare
• Depth is okay

Status : 14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 4

Basic Paradigmas of modelling Document
based Databases
• No/little normalization
• Duplication is acceptable „embrace duplication“
• Joins/References are rare
• Depth is okay (=Arrays, embedded objects and combinations of both)
• Reading operations are „cheap“ (fast), updating operations and joins are
expensive (slow)
• Difference between static and dynamic properties is important
• While modelling think in big scale
• Joins are bottlenecks but better than frequent update processes on large scale
• Work backwards = from use case to data model
• Are there relationships (1:n, m:n)
Status : 14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 5
Basic Paradigmas of modelling Document
based Databases
• No/little normalization
• Duplication is acceptable „embrace duplication“
• Joins are rare
• Depth is okay (=Arrays, embedded objects and combinations of both)
• Duplication is okay (f.ex. To make queries faster)
• Reading operations are „cheap“ (fast), updating operations and joins are
expensive (slow)
• Difference between static and dynamic properties is important
• While modelling think in big scale
• Joins are bottlenecks but better than frequent update processes on large scale
• Work backwards = from use case to data model
• „One query for one page“-objective (or at least very little queries)
Status : 14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 6
Reasons for external References/FK
• Multiple documents reference the same object
• Object has some volatility → updating would be necassary for all
these references → would be very expensive
• Embedded (reduced) information should be accesible for deeper look
on data (drill-in)

Status : 14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 7

Solutions for 1:n-Relations
• Embedding in an array:
• Could become giant (16MB-limit)
• Extract subdocument to proper collection and reference it bei
$lookup
• At bigger scale very „expensive“
• Summary fields to prefetch a certain subset and reference for details

Status : 14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 8

Handeling versions
• In case of changing requirements changes are much easier than on
relational DB/DDL-structure.
• Recomandations:
• Add version of „schema“ to handle different fields on the backend
• use $jsonschema to define requirements that are absolutely necassary

Status : 14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 9

Enforcing datamodel by using $jsonschema
db.createCollection("students", {
validator: {
$jsonSchema: {
bsonType: "object",
title: "Student Object Validation",
required: [ "address", "major", "name", "year" ],
properties: {
name: {
bsonType: "string",
description: "'name' must be a string and is required"
},…} )#

https://www.mongodb.com/docs/manual/core/schema-validation/specify-json-schema/#std-label-schema-validation-json
Status : 14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 10
$jsonschema enforcing certain values
db.createCollection("shipping", {
validator: {
$jsonSchema: {
bsonType: "object",
title: "Shipping Country Validation",
properties: {
country: {
enum: [ "France", "United Kingdom", "United States" ],
description: "Must be either France, United Kingdom, or
United States" } } } }} )

https://www.mongodb.com/docs/manual/core/schema-validation/specify-json-schema/specify-allowed-field-values/#std-label-schema-allowed-field-values

Status : 14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 11

Take Care!
• Document based databases are not slimmer (many redundancies)
• For fundamental noSQL-benefits (speed, cost, scalability) they need
as much or even more pre-definitions and modelling
• Prepare and model your most important „views“ beforehand
• Plan scalibility by modelling logical and therefore partitionable units (classes)
• Know the focus of your application (write-load, read-load)

NoSQL is not a „quick“ substitue for relational

databases!!

Status : 14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 12

Guidelines embedding versus referencing
• Many reading processes
• Reading processes are frequently
in the combination of super- and
sub-document
• Sub-documents have „natural“
limit („one-to-few“-relations)
• Many static data sets/entities
EMBEDDING

14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 13

Guidelines embedding versus referencing
• Many writing processes →
• No practical limit to document
size (f.ex. unlimmited
commentary possibilites on
posting) →

REFERENCING

14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 14

Use Cases

14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 15

Use Case „E-Commerce Website“
• Use Case Wish List
• Use Case Shopping Card
• Use Case Order History

Status : 14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 16

Assignment1: Evaluation of Use Cases
1) Define the objects involved in this use case with their attributes
• Class definition (OOP) OR
• Highlight and summarize them on print out
2) Which of the data do you consider to be
• Static
• Dynamic

Means of modelling: object definition (1), arrays, embedded documents

Status : 14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 17

A1 Sample:
Timetable per class
1. Class: perspective of view →
json-file „class“
2. Headmaster: (very likely)
static | single value →
Attribute of class
3. Week: static | array of values
→ embedding into class
4. Sessions per hour: (most
times) static | limited amount
of informational unit „lesson“
with attributes name, teacher,
room, further specifications
→ embedded object per day
14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 18
A1 Sample:
Timetable sample json
1. File per class
2. Embedded teacher (object)
3. Many lessons per teacher → sessions could
be embedded into teacher
4. Sessions are repeated per week, so weekday
could be embedded into sessions

14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 19

Assignment2: Defining Schema
1) Define a propriate json-Schema for your use-case
2) Implement the schema on your mongoDB
3) Model a class covering the needs of your e-Formular on your
MongoProject (Java, C#, …)
4) Test your schema by inserting valid and invalid test data

5) Bonus: develop a secondary version of your schema and define a

possible handling of it on a query example.

Status : 14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 20

A2 Sample:
$jsonschema
{ %%,
"$schema": "http://json-schema.org/draft-04/schema#", "Lehrer": { "type": "array",
"type": "array", "items": [ {
"items": [{ "type": "object",
"type": "object", "properties": {
"properties": { "Id": { "type": "string" },
"Id": { "type": "string" }, "Zuname": { "type": "string"},
"Kv": { "type": "object", "Vorname": { "type": "string" },
"properties": { "Gegenstaende": { "type": "array",
"Id": { "type": "string}, "items": [ { "type": "object",
"Zuname": { "type": "string" }, "properties": {"Id": {
"Vorname": { "type": "string“ } }, "type": "string" }, "Name": {
"required": [ "Id", "Zuname", "Vorname" ] "type": "string"
%% }, "Zeiten": { "type": "array",
"items": [ { "type": "object", "properties": {

14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 21

Assignment3: Introducing Referencing
1) Identify drivers for referencing
1) (potentially) dynamic datasets
2) write-heavy processes
3) Unlimited arrays

2) Alter the model to (partially) referenced structures

Status : 14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 22

E-Formular: Eignungstest

Status : 14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 23

E-Formular: Abschlussprüfungen

Status : 14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 24

E-Formular: Schulbesuchsbestätigungen

Status : 14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 25

E-Formular: Wahlpflichtfächer

Status : 14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 26

E-Formular: Religionsanmeldung

Status : 14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 27

E-Formular: Zeugnis/Aufstieg

Status : 14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 28

E-Formular: Zeugnis/Nachprüfung

Status : 14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 29

E-Formular: Sommerschule

Status : 14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 30

Sources
• John Cokos:
https://www.linkedin.com/learning/data-
modeling-in-mongodb/introduction-to-
data-
modeling?autoplay=true&u=37766780
• E-Formular: e-formular.spengergasse.at
• MongoDB-Doku:
https://www.mongodb.com
• https://learn.microsoft.com/en-
us/azure/cosmos-db/nosql/modeling-
data

"Dieses Foto" von Unbekannter Autor ist lizenziert gemäß CC BY-SA

14.12.2022 DI(FH) Mag. Johanna Maria NIKLAS 31