How do I practice coding and algorithm questions?

Use PracHub's coding console to write, test, and debug your solutions in Python or JavaScript. View hints, test against sample inputs, and compare with official solutions.

What difficulty level is this coding question?

This is a medium difficulty Coding & Algorithms question, commonly asked during Take-home Project rounds at Stripe.

What role is this question designed for?

This question is commonly asked for Software Engineer candidates at Stripe during technical interviews.

Implement a Datacenter Request Router | Stripe Coding Question

Implement a Datacenter Request Router

Company: Stripe

Role: Software Engineer

Category: Coding & Algorithms

Difficulty: medium

Interview Round: Take-home Project

Implement a command processor for a request-routing system that manages datacenters. Each datacenter stores: - `name` - `latitude` in `[-90, 90]` - `longitude` in `[-180, 180]` - `capacity`, which must be greater than `0` - `healthy` status, initially `true` - `current_load`, initially `0` Process commands sequentially and return one output per command. ## Commands ### 1. `REGISTER <name> <lat> <lon> <capacity>` Create a new datacenter. - Reject the command if: - latitude is outside `[-90, 90]` - longitude is outside `[-180, 180]` - capacity is `<= 0` - the datacenter name already exists - Output `OK` on success, otherwise output `ERROR`. ### 2. `SET_HEALTHY <name> <true|false>` Update the health status of an existing datacenter. - Output `OK` on success. - Output `ERROR` if the datacenter does not exist. ### 3. `DISTANCE <lat1> <lon1> <lat2> <lon2>` Compute the great-circle distance between two coordinates. - Validate both points using the same latitude/longitude bounds as above. - Use the Haversine formula with Earth radius `6371 km`. - Output the distance rounded to the nearest whole kilometer. - If any coordinate is invalid, output `ERROR`. ### 4. `ROUTE <userLat> <userLon>` Route one user request to a datacenter. - Consider only healthy datacenters. - Compute the distance from the user to every healthy datacenter. - Sort candidates by distance ascending. - If two datacenters have the same distance, sort by name in lexicographic order. - Scan the sorted list and choose the first datacenter whose `current_load < capacity`. - If a datacenter is selected: - increment its load by `1` - output: `<selected_name> <selected_distance> <comma-separated healthy datacenter names in sorted order>` - If all healthy datacenters are full, output: `None <comma-separated healthy datacenter names in sorted order>` - If there are no healthy datacenters, output `None`. ## Example 1 Input commands: - `REGISTER us-west 38 -122 100` - `REGISTER us-east 41 -74 150` - `REGISTER us-west 50 -100 50` - `REGISTER invalid-node 91 0 100` - `REGISTER invalid-cap 0 0 0` - `SET_HEALTHY us-east false` - `SET_HEALTHY fake-node true` Output: - `OK` - `OK` - `ERROR` - `ERROR` - `ERROR` - `OK` - `ERROR` ## Example 2 Input commands: - `DISTANCE 38 -122 41 -74` - `DISTANCE 0 0 0 0` - `DISTANCE 91 0 0 0` Output: - `4080` - `0` - `ERROR` ## Example 3 Input commands: - `REGISTER node-A 0 0 1` - `REGISTER node-B 0 0 1` - `REGISTER node-C 10 10 100` - `SET_HEALTHY node-C false` - `ROUTE 0 0` - `ROUTE 0 0` - `ROUTE 0 0` Output: - `OK` - `OK` - `OK` - `OK` - `node-A 0 node-A,node-B` - `node-B 0 node-A,node-B` - `None node-A,node-B`

Quick Answer: This question evaluates command parsing, stateful data structure management, input validation, geospatial computation using the Haversine formula, deterministic sorting/tie-breaking, and capacity-based routing logic.

Implement a command processor for a request-routing system that manages datacenters. Write a function `solution(commands)` that processes the commands in order and returns a list of output strings, one output per command. Each datacenter stores: - `name` - `latitude` in `[-90, 90]` - `longitude` in `[-180, 180]` - `capacity` as a positive integer - `healthy` status, initially `true` - `current_load`, initially `0` Supported commands: 1. `REGISTER <name> <lat> <lon> <capacity>` - Create a new datacenter. - Reject the command and output `ERROR` if: - latitude is outside `[-90, 90]` - longitude is outside `[-180, 180]` - capacity is `<= 0` - the datacenter name already exists - Otherwise output `OK`. 2. `SET_HEALTHY <name> <true|false>` - Update the health status of an existing datacenter. - Output `OK` on success. - Output `ERROR` if the datacenter does not exist. 3. `DISTANCE <lat1> <lon1> <lat2> <lon2>` - Compute the great-circle distance between two coordinates using the Haversine formula. - Use Earth radius `6371 km`. - Validate both points using the same latitude/longitude bounds as above. - Output the distance rounded to the nearest whole kilometer. - If any coordinate is invalid, output `ERROR`. 4. `ROUTE <userLat> <userLon>` - Validate the user coordinates using the same latitude/longitude bounds. If invalid, output `ERROR`. - Consider only healthy datacenters. - Compute the distance from the user to every healthy datacenter. - Sort healthy datacenters by exact computed distance ascending. If two distances are equal, sort by datacenter name in lexicographic order. - Scan that sorted list and choose the first datacenter whose `current_load < capacity`. - If a datacenter is selected: - increment its `current_load` by `1` - output `<selected_name> <selected_distance> <comma-separated healthy datacenter names in sorted order>` - `selected_distance` must be rounded to the nearest whole kilometer - If all healthy datacenters are full, output `None <comma-separated healthy datacenter names in sorted order>`. - If there are no healthy datacenters, output `None`. Notes: - The comma-separated list in `ROUTE` must follow the same sorted order used for routing. - Datacenter loads persist across future commands.

Constraints

0 <= len(commands) <= 2000
Datacenter names are non-empty strings without spaces
Latitude and longitude values may be integers or decimals
Capacity is intended to be an integer and must be greater than 0
At most 2000 datacenters will be registered

Examples

Input: []

Expected Output: []

Explanation: No commands means no outputs.

Input: ["REGISTER us-west 38 -122 2", "REGISTER us-east 41 -74 3", "REGISTER us-west 50 -100 1", "REGISTER invalid-node 91 0 100", "REGISTER invalid-cap 0 0 0", "SET_HEALTHY us-east false", "SET_HEALTHY fake-node true"]

Expected Output: ["OK", "OK", "ERROR", "ERROR", "ERROR", "OK", "ERROR"]

Explanation: This checks successful registration, duplicate names, invalid coordinates, invalid capacity, updating an existing datacenter, and failing to update a missing one.

Input: ["DISTANCE 0 0 0 0", "DISTANCE 0 0 90 0", "DISTANCE 0 0 0 180", "DISTANCE 91 0 0 0"]

Expected Output: ["0", "10008", "20015", "ERROR"]

Explanation: Same point is 0 km, equator to pole is a quarter of Earth's circumference, opposite points on the equator are half the circumference, and invalid coordinates produce ERROR.

Input: ["REGISTER node-B 0 0 1", "REGISTER node-A 0 0 1", "REGISTER node-C 10 10 100", "SET_HEALTHY node-C false", "ROUTE 0 0", "ROUTE 0 0", "ROUTE 0 0"]

Expected Output: ["OK", "OK", "OK", "OK", "node-A 0 node-A,node-B", "node-B 0 node-A,node-B", "None node-A,node-B"]

Explanation: node-A and node-B are tied at distance 0, so lexicographic order breaks the tie. After both reach capacity 1, routing returns None along with the healthy sorted list.

Input: ["REGISTER gamma 0 0 1", "REGISTER alpha 0 180 1", "REGISTER beta 90 0 1", "ROUTE 0 0", "ROUTE 0 0", "ROUTE 95 0", "SET_HEALTHY gamma false", "ROUTE 0 0"]

Expected Output: ["OK", "OK", "OK", "gamma 0 gamma,beta,alpha", "beta 10008 gamma,beta,alpha", "ERROR", "OK", "alpha 20015 beta,alpha"]

Explanation: The first route chooses the closest node, the second skips a full but still healthy node, invalid user coordinates return ERROR, and after gamma is marked unhealthy only beta and alpha remain in sorted order.

Input: ["REGISTER solo 0 0 1", "SET_HEALTHY solo false", "ROUTE 0 0"]

Expected Output: ["OK", "OK", "None"]

Explanation: When there are no healthy datacenters, ROUTE returns exactly None.

Hints

Use a hash map keyed by datacenter name so REGISTER, duplicate checks, and SET_HEALTHY updates are O(1).
For each ROUTE command, build a list of tuples like (exact_distance, name), sort it once, and then scan for the first datacenter that still has remaining capacity.

Quick Overview

This question evaluates command parsing, stateful data structure management, input validation, geospatial computation using the Haversine formula, deterministic sorting/tie-breaking, and capacity-based routing logic.