Media storage and CDN in HLD - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Architecture

intermediate

Design a scalable media storage system for millions of users

You need to design a media storage system that can handle millions of users uploading images and videos daily. Which architectural component is most critical to ensure scalability and fast access?

A Store media files in distributed object storage and metadata in a separate database.

B Store media files directly in the application server's local disk for faster access.

C Use a peer-to-peer network where users share media files directly without central storage.

D Use a single large database server to store all media files and metadata.

💡 Hint

Think about separating file storage from metadata and how to handle large scale.

Distributed object storage allows scalable, durable storage of large media files. Metadata is stored separately for efficient querying. This separation supports scalability and performance.

❓ scaling

intermediate

Handling sudden traffic spikes in CDN delivery

Your CDN is delivering media content globally. Suddenly, a viral video causes a huge spike in traffic. What is the best way to handle this spike without degrading user experience?

A Disable caching so all requests go to the origin server for fresh content.

B Increase the origin server capacity to handle all requests directly.

C Rely on CDN edge caching to serve most requests and scale edge nodes automatically.

D Limit user access by blocking some regions temporarily to reduce load.

💡 Hint

Consider how CDNs reduce load on origin servers during high traffic.

CDN edge caching serves content close to users, reducing origin load. Automatic scaling of edge nodes helps handle spikes smoothly.

❓ tradeoff

advanced

Choosing between CDN push and pull models

When designing a CDN for media delivery, you must choose between push and pull models. Which tradeoff is true?

A Pull CDN preloads all content to edges, which reduces storage cost but increases latency.

B Push CDN requires preloading content to edge servers, increasing storage cost but reducing first-request latency.

C Push CDN always delivers content slower than pull CDN because of extra replication steps.

D Pull CDN requires manual upload of content to each edge server before delivery.

💡 Hint

Think about when content is sent to edge servers in each model.

Push CDN sends content to edges ahead of time, increasing storage but reducing latency on first request. Pull CDN fetches content on demand, saving storage but adding latency initially.

❓ component

advanced

Key components of a media CDN architecture

Which component is essential in a CDN architecture to ensure media content is delivered efficiently and securely?

A Edge servers that cache media content close to users and handle most delivery requests.

B Origin server that stores the original media files and handles all user requests.

C Load balancer that distributes user requests only among origin servers.

D Database cluster that stores user credentials and media metadata but does not cache content.

💡 Hint

Consider which component reduces latency and bandwidth usage.

Edge servers cache content near users, reducing latency and origin load. They are critical for efficient CDN delivery.

❓ estimation

expert

Estimating storage and bandwidth for a global media CDN

Your service expects 10 million daily active users uploading 5 MB average media files each, and 100 million daily views with average 3 MB per view. Estimate the daily storage and bandwidth requirements for the CDN.

A Storage: 50 TB per day; Bandwidth: 30 TB per day

B Storage: 500 TB per day; Bandwidth: 30 TB per day

C Storage: 5 TB per day; Bandwidth: 3000 TB per day

D Storage: 50 TB per day; Bandwidth: 300 TB per day

💡 Hint

Calculate storage from uploads and bandwidth from views multiplied by average sizes.

Storage = 10M users * 5 MB = 50 TB daily. Bandwidth = 100M views * 3 MB = 300 TB daily.