System requirements

Functional:

List functional requirements for the system (Ask the chat bot for hints if stuck.)...

UI

• Login
• Enter a URL and get a short URL
• Must be unique
• Avoid characters that could be confused
  • Alway lowercase, eliminate numbers that could be confused with letters
    • l and 1
    • o and 0
    • g and 9
• View list of short urls
  • Statistics
    • Clicks/time
    • Clicks/referrers
• Delete a short url
• Update long url

Service

• Short URL redirect to the long URL

Track analytics

• date/time
• count
• referrer

Non-Functional:

List non-functional requirements for the system...

Performance: TBD requests/second

Reliability: Aim for 100% uptime, deployment should not cause downtime

Scalability: Automatic scaling to handle traffic spikes

Capacity estimation

Estimate the scale of the system you are going to design...

• How many users?
  • How many concurrently using the service?
• How many urls/user?
• How many clicks on a short url?
  • How many concurrent?

API design

Define what APIs are expected from the system...

Create short URL

Endpoint: POST /short

Authentication: Session

Body:

{

"url": ""

}

Response:

{

"shortUrl": ""

}

Get short URLs

Endpoint: GET /short?max={max}&offset={offset}

Params:

• max - Number of urls to return
• offset - Paging offset

Response:

[

{

shortUrl: "",

longUrl: ""

},

...

]

Redirect short URL to long URL

Endpoint: GET /{short-code}

Response: Redirects to the long URL

Database design

Defining the system data model early on will clarify how data will flow among different components of the system. Also you could draw an ER diagram using the diagramming tool to enhance your design...

Table: url_mapping

Columns:

id - unique id (auto-generated primary key or other unique id, e.g. ObjectId if Mongo)

customer_id - foreign key to the customer

short_code - the unique short code to go at the end of the url

url - the original url to redirect to

dateCreated - date the record was created

lastUpdated - date the record was last modified

Table: url_metric

id - unique id

short_code - the short code

metric_type_id - foreign key to the metric table

value - metric value

date - date of the metric

Table: metric_type

id - unique id

name - metric name

type -

Table: customer

id

email

High-level design

You should identify enough components that are needed to solve the actual problem from end to end. Also remember to draw a block diagram using the diagramming tool to augment your design. If you are unfamiliar with the tool, you can simply describe your design to the chat bot and ask it to generate a starter diagram for you to modify...

Components

• Front-end application
  • Login/Create account
  • Manage urls (CRUD)
• Back-end API
  • REST endpoints for CRUD
  • redirect url for short code to long url
• DB
  • User accounts
  • URL to short code mapping
  • URL usage metrics

Request flows

Explain how the request flows from end to end in your high level design. Also you could draw a sequence diagram using the diagramming tool to enhance your explanation...

Customer -> Create account/login

Customer -> Create short code for url

Service -> Return short code url to customer

End user -> Clicks short code url

Service -> Redirects to original url and tracks the click

Detailed component design

Dig deeper into 2-3 components and explain in detail how they work. For example, how well does each component scale? Any relevant algorithm or data structure you like to use for a component? Also you could draw a diagram using the diagramming tool to enhance your design...

Front-end:

Choose a popular and well-architected framework, such as React, Vue, Angular, or Svelte.

Back-end:

For the CRUD APIs choice of technology doesn't matter too much, any framework that can produce APIs should do. But, for the redirect url we'll probably want to go with something that is very low-latency and scales horizontally. Traffic will likely be quite variable as well. A CDN in front of a serverless (e.g. AWS Lambda) endpoint will scale very well. Cloudfront metrics could be used to power the usage analytics feature.

We will need a service to generate short codes.

Possibilities:

1) Take the primary key of the short code record and generating a BaseX encoding.

• Pro: no need to check for collisions
• Con: can be iterated by attackers

2) Generate a random string of X characters/numbers. Maybe start with 4 and then grow it after we've exhausted the sample space.

• Pro: non-sequential makes it less vulnerable to attack
• Con: need to check for collisions when creating or create a set of possible values up front to choose from

Trade offs/Tech choices

Explain any trade offs you have made and why you made certain tech choices...

Failure scenarios/bottlenecks

Try to discuss as many failure scenarios/bottlenecks as possible.

Future improvements

What are some future improvements you would make? How would you mitigate the failure scenario(s) you described above?