[mattLummus]

Is it possible that a regular time series could have better read performance (particularly in aggregations) vs an irregular one due to determinism/randomness - or is that irrelevant to the underlying implementation?