User, Group, Expense classes in LLD - Practice Problems & Coding Challenges

Challenge - 5 Problems

🧠 Conceptual

intermediate

Identify the correct relationship between User, Group, and Expense classes

In a system where users can belong to groups and share expenses, which of the following best describes the relationship between User, Group, and Expense classes?

A Users belong to a single Group only; Expenses belong to Users only; Groups do not track Expenses.

B A User can belong to multiple Groups; a Group contains multiple Users; an Expense is linked to a Group and paid by a User.

C An Expense belongs to multiple Users; Groups are independent and do not contain Users; Users pay Expenses individually.

D A Group belongs to a single User; an Expense belongs to a User only; Users do not belong to Groups.

💡 Hint

Think about how users collaborate in groups and how expenses are shared.

Users can be members of multiple groups, groups contain multiple users, and expenses are linked to groups and paid by users.

❓ Architecture

intermediate

Choose the best class attribute to represent an Expense's payer

Which attribute best represents the payer of an Expense in the Expense class?

A payer_group: stores the Group object that paid the expense.

B payer_name: stores the name of the User who paid the expense as a string.

C payer_id: stores the unique identifier of the User who paid the expense.

D payer_amount: stores the amount paid by the User.

💡 Hint

Consider how to uniquely identify the payer without duplicating user data.

Using payer_id links the expense to the User uniquely and avoids data duplication.

❓ scaling

advanced

Scaling the Group class for thousands of Users

When a Group contains thousands of Users and hundreds of Expenses, which design choice helps maintain performance and scalability?

A Store only User IDs in the Group class and fetch User details on demand.

B Store User objects directly inside the Group class as a list.

C Duplicate User data inside each Group to avoid fetching from User class.

D Store Expense objects inside each User instead of the Group.

💡 Hint

Think about memory usage and data duplication when handling large numbers.

Storing only User IDs reduces memory usage and allows lazy loading of User details, improving scalability.

❓ tradeoff

advanced

Tradeoff between storing Expense split details in Expense vs User

Where should the details of how an Expense is split among Users be stored for best design?

A Store split details in a separate Split class linking Users and Expenses.

B Store split details inside each User class as a list of Expenses and amounts.

C Store split details inside the Expense class as a map of User IDs to amounts.

D Do not store split details; calculate splits dynamically each time.

💡 Hint

Consider separation of concerns and flexibility for future changes.

A separate Split class cleanly models the many-to-many relationship and allows flexible queries.

❓ estimation

expert

Estimate storage for 1 million Users, 100k Groups, and 10 million Expenses

Estimate the approximate storage needed if each User object requires 1KB, each Group object 2KB, and each Expense object 0.5KB. Ignore overhead and indexes.

A Approximately 8.5 GB

B Approximately 7.5 GB

C Approximately 5.5 GB

D Approximately 6.5 GB

💡 Hint

Calculate total size by multiplying counts by sizes and summing.

Users: 1,000,000 * 1KB = 1,000,000 KB = ~1GB
Groups: 100,000 * 2KB = 200,000 KB = ~0.2GB
Expenses: 10,000,000 * 0.5KB = 5,000,000 KB = ~5GB
Total ~6.2GB rounded to 6.5GB.