RateLimiting

RateLimiting sampler

This sampler tries to sample with a rate per time window, which by default is 0.1 traces/second. When applied to a child Span of a sampled parent Span, the child Span keeps the sampling decision.

Implementation details

RateLimiting sampling aims to solves some problems:

1. Getting QPS based sampling
2. Providing real sampling probabilities
3. Minimal overhead

To achieve rate-limiting, the time that we last made a QPS based sampling decision is stored in an atomic variable. The elapsed time Z since we last made a probabilistic decision is also noted. We then use a probability function P(Z) such that we get the desired sampling QPS. We always want P(Z) to be very cheap to compute.

Thus if X is the desired QPS, Z is the elapsed time in seconds, since the last sampling decision, then

P(Z) = min(Z * X, 1)

References