MyDotey.QuantileEstimator 1.0.2

Quantile Estimator

Commmon Library collecting multiple quantile algorithms for streaming data, concurrent & easy to use.

Implemented in java/dotnet separately. Code is clean. Easy to read.

Usage

• java
  https://github.com/mydotey/quantile-estimator/tree/master/java

• dotnet
  https://github.com/mydotey/quantile-estimator/tree/master/dotnet

Features

• Thread Safe

  • add value: lock free

  • batch get quantiles: synchronized

• Time Window

  • better accuracy by time window rotate

Algorithms

• Classic Algorithm

  • accurate

  • no data compaction

• CKMS Algorithm

  • data compaction & space efficient

  • predefined quantile error

  • most used approximate algorithm

• GK Algorithm

  • data compaction & space efficient

• KLL Algorithm

  • data compaction & space efficient

  • new algorithm

Papers

• Cormode, Korn, Muthukrishnan, and Srivastava. "Effective Computation of Biased Quantiles over Data Streams" in ICDE 2005

• Greenwald and Khanna. "Space-efficient online computation of quantile summaries" in SIGMOD 2001

• Zohar Karnin, Kevin Lang and Edo Liberty. "Optimal Quantile Approximation in Streams" in FOCS 2016

Others' Projects

• https://github.com/umbrant/QuantileEstimation

• https://github.com/edoliberty/streaming-quantiles

• https://github.com/tdunning/t-digest

Developers

• Qiang Zhao <koqizhao@outlook.com>