The interview was conducted on Google Docs and lasted 45 minutes.

The interviewer asked the following problem:

Distributed Communication Startup Problem

Google operates a distributed service consisting of multiple processes.

During startup, communication channels between processes become available gradually as different subsystems initialize. Once a communication channel becomes available, it remains available for the rest of the startup sequence.

A process can send a message to another process if there exists a chain of available communication channels connecting them.

Given the startup log and two process IDs, determine the earliest time at which the source process can successfully send a message to the destination process.

You are given:

n processes numbered from 0 to n-1

m communication channels

Each communication channel is represented as:

(processA, processB, activationTime)

meaning that beginning at activationTime, the two processes can communicate directly.

You are also given:

source destination

Return the earliest timestamp at which a message sent from source can reach destination.

Example:

n = 6

channels =

(0,1,2) (1,2,5) (0,3,3) (3,4,6) (4,5,7) (2,5,8)

source = 0 destination = 5

answer 7

( answer was not directly given i have to do the dry run of my code to arrive at that).

No constraints were given initially. The interviewer gradually revealed more details as the discussion progressed. I first discussed the brute-force approach and then moved toward the optimal solution.

I eventually arrived at the correct logic, but it happened quite late in the interview, and I wasn't able to finish coding the complete solution in time. Because of that, I wasn't able to make it through the process.

Still, it was a great learning experience.

All the best to everyone appearing for the process. Have fun, enjoy the experience, and make the most of the opportunity!