mayankshriv commented on a change in pull request #5118: Synthetic Time Series Generator for pinot-tools URL: https://github.com/apache/incubator-pinot/pull/5118#discussion_r388627944
########## File path: pinot-tools/src/main/java/org/apache/pinot/tools/data/generator/TemplateSpikeGenerator.java ########## @@ -0,0 +1,108 @@ +/** + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ +package org.apache.pinot.tools.data.generator; + +import org.apache.commons.math3.distribution.LogNormalDistribution; + +import java.util.Map; + +/** + * TemplateSpikeGenerator produces a series of log-normal spikes with log-normal arrival times, with optional smoothing. + * This pattern is typical for rare even spikes, such as error counts. The generated values are sampled non-deterministically. + * + * Generator example: + * <pre> + * baseline = 0 + * arrivalMean = ? + * magnitudeMean = ? + * + * returns [ 0, 0, 0, 0, 0, 0, 47, 15, 2, 1, 0, 0, ... ] + * </pre> + * + * Configuration examples: + * <ul> + * <li>./pinot-tools/src/main/resources/generator/simpleWebsite_generator.json</li> + * <li>./pinot-tools/src/main/resources/generator/complexWebsite_generator.json</li> + * </ul> + */ +public class TemplateSpikeGenerator implements Generator { + private final double baseline; + private final double smoothing; + + private final LogNormalDistribution arrivalGenerator; + private final LogNormalDistribution magnitudeGenerator; + + private long step = -1; + + private long nextArrival; + private double lastValue; + + public TemplateSpikeGenerator(Map<String, Object> templateConfig) { + this(toDouble(templateConfig.get("baseline"), 0), + toDouble(templateConfig.get("arrivalMean"), 2), + toDouble(templateConfig.get("arrivalSigma"), 1), + toDouble(templateConfig.get("magnitudeMean"), 2), + toDouble(templateConfig.get("magnitudeSigma"), 1), + toDouble(templateConfig.get("smoothing"), 0)); + } + + public TemplateSpikeGenerator(double baseline, double arrivalMean, double arrivalSigma, double magnitudeMean, double magnitudeSigma, double smoothing) { + this.baseline = baseline; + this.smoothing = smoothing; + + this.arrivalGenerator = new LogNormalDistribution(arrivalMean, arrivalSigma); + this.magnitudeGenerator = new LogNormalDistribution(magnitudeMean, magnitudeSigma); + + this.nextArrival = (long) arrivalGenerator.sample(); + this.lastValue = baseline; + } + + @Override + public void init() { + // left blank + } + + @Override + public Object next() { + step++; + + if (step < nextArrival) { + lastValue = (1 - smoothing) * baseline + smoothing * lastValue; + return (long) lastValue; + } + + nextArrival += (long) arrivalGenerator.sample(); + lastValue = baseline + this.magnitudeGenerator.sample(); + return (long) lastValue; + } + + private static double toDouble(Object obj, double defaultValue) { + if (obj == null) { + return defaultValue; + } + return Double.valueOf(obj.toString()); + } + + public static void main(String[] args) { Review comment: Is this needed? ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
