Compute 5-minute rolling average load

You are building a monitoring component for a key–value (KV) store. Each request contributes 1 unit of load at its request time.

Design a data structure that supports the following operations efficiently:

• record(t) : a request arrives at integer timestamp t (in seconds). Timestamps are non-decreasing across calls.
• avgLoad(t) : return the average load per second over the last 5 minutes ending at time t , i.e., over the interval (t - 300, t] .

Requirements

• Treat the time window as exactly 300 seconds .
• If there are no requests in the window, return 0 .
• avgLoad(t) should return a floating-point value:

$\text{avg} = \frac{\#\text{requests in }(t-300, t]}{300}$

Example

If requests were recorded at times 1, 2, 300, 301:

• avgLoad(301) = 4 / 300
• avgLoad(600) = 0 / 300 = 0

Constraints / Notes

• The system may receive a large number of requests, so avgLoad(t) should be faster than scanning all historical events.
• You may assume single-threaded access (no concurrency concerns).