Proximity Request Routing: Route Incoming Requests to the Nearest Available Server

Question

## Problem

You have `n` servers on a 2D grid, each at a position `(x, y)` with a maximum capacity `c`. Given a list of incoming requests, each at a position `(rx, ry)`, route each request to the nearest server (Manhattan distance) that still has remaining capacity. If no server has capacity, return -1 for that request.

```python
from typing import List

def route_requests(
    servers: List[List[int]],
    requests: List[List[int]]
) -> List[int]:
    # servers: [[x, y, capacity], ...] (0-indexed ids)
    # requests: [[rx, ry], ...]
    # return: [server_id or -1, ...]
    pass
```

**Example:**
```
servers  = [[0,0,2],[5,0,1]]
requests = [[1,0],[1,0],[1,0]]

Req 0: closest=server0 (dist=1), assign -> cap=[1,1]
Req 1: closest=server0 (dist=1), assign -> cap=[0,1]
Req 2: server0 full; next=server1 (dist=4), assign -> cap=[0,0]

Output: [0, 0, 1]
```

## Follow-ups

1. How would you use a priority queue to efficiently find the nearest available server for each request?
2. What if servers can also decline requests based on request type — how do you model that?
3. In a geographically distributed system, how does Euclidean/Manhattan distance translate to real-world routing latency?
4. How would you handle server failures mid-processing and reroute in-flight requests?

Proximity Request Routing: Route Incoming Requests to the Nearest Available Server

Question Details

Full Details

Topics

More from Stripe