Answer: Count the number of home page requests that load in under 100 ms, and then divide by the total number of home page requests.

The question asks for the Google-recommended way to calculate a Service Level Indicator (SLI) for home page request latency with an acceptable threshold of 100 ms. In Google's Site Reliability Engineering (SRE) practices, an SLI for latency is typically defined as the proportion of requests that meet a specific performance threshold. This involves counting the number of successful requests (those faster than the threshold) and dividing by the total relevant requests to measure compliance. Option C correctly implements this by focusing only on home page requests and calculating the ratio of requests under 100 ms to the total home page requests. This aligns with Google's guidance, such as in the SRE book, which emphasizes using a threshold-based success rate for latency SLIs. Option A is incorrect because computing a percentile at a fixed value (100 ms) does not directly measure the proportion of requests meeting the threshold. Option B is not specific to the 100 ms threshold and instead computes percentiles, which are more suited for distribution analysis than a binary SLI. Option D is incorrect because it dilutes the metric by including all web application requests, which may include non-home-page traffic and misrepresents the home page's performance.

Author: LeetQuiz Editorial Team