Tyler Yahn

> `Exemplar` sampling SHOULD be turned on by default. We do not comply with this.

> If `Exemplar` sampling is off, the SDK MUST NOT have overhead related to exemplar sampling. Our current implementation complies with this.

> A Metric SDK MUST allow exemplar sampling to leverage the configuration of > metric aggregation. For example, Exemplar sampling of histograms should be able > to leverage bucket boundaries....

> The `ExemplarFilter` configuration MUST allow users to select between one of the > built-in ExemplarFilters. While `ExemplarFilter` determines which measurements > are _eligible_ for becoming an `Exemplar`, the `ExemplarReservoir`...

> An OpenTelemetry SDK MUST support the following filters: > > - AlwaysOn > - AlwaysOff > - TraceBased Our implementation supports these filters.

> The `ExemplarReservoir` interface MUST provide a method to offer measurements to the reservoir and another to collect accumulated Exemplars. https://github.com/open-telemetry/opentelemetry-go/blob/7ee6ff19b51eb4bffdd48639ac5698c9ee8932d6/sdk/metric/internal/exemplar/reservoir.go#L14-L33 We comply with this.

> A new `ExemplarReservoir` MUST be created for every known timeseries data point, as determined by aggregation and view configuration. This data point, and its set of defining attributes, are...

> The "offer" method SHOULD have the ability to pull associated trace and span > information without needing to record full context. In other words, current > span context and...

> The SDK MUST include two types of built-in exemplar reservoirs: > > 1. `SimpleFixedSizeExemplarReservoir` > 2. `AlignedHistogramBucketExemplarReservoir` > > By default: > > - Explicit bucket histogram aggregation with...

> #### SimpleFixedSizeExemplarReservoir > > This reservoir MUST use an uniformly-weighted sampling algorithm based on the > number of samples the reservoir has seen so far to determine if the...