Improve Summary quantiles with DataSketches

Summary.observe() can become expensive when quantiles are recorded at high frequency. This can make the current quantile path visible on hot request paths.

We saw this in ZooKeeper's Prometheus metrics path. In an internal ZooKeeper 3.9.2 fork, a version inspired by ZooKeeper's unmerged DataSketches Summary PR improved peak throughput by about 2x.

DataSketches KLL may be a useful way to improve this in client_java. The goal would be to reduce the cost of the observe path while keeping the external Summary behavior as close as practical.

This would not have to replace the current CKMS-based Summary immediately. DataSketches KLL has a different accuracy model, memory cost, and quantile visibility behavior, so an explicit opt-in path may be a better first step.

Initial questions:

• Does using DataSketches for Summary quantiles sound like a direction worth exploring?
• If so, would a separate opt-in artifact be a reasonable way to introduce it?
• What behavior details and benchmark data would be most useful before going further?

Beginne bei Summary.observe() und verfolge den aktuellen CKMS-basierten Quantilpfad in client_java. Prüfe den ZooKeeper DataSketches Summary PR und vergleiche einen KLL-basierten Ansatz anhand von Benchmark-Daten, einschließlich Genauigkeit, Speicherverbrauch und Quantil-Sichtbarkeit. Als abgeschlossen gilt die Arbeit, wenn eine dokumentierte Richtung und Belege dafür vorliegen, ob sich eine Opt-in-Implementierung lohnt.