VictoriaMetrics/docs/vmanomaly.md

sort	weight	title	menu	aliases
11	11	vmanomaly	docs parent weight victoriametrics 11	/vmanomaly.html

vmanomaly

vmanomaly is a part of enterprise package. Please contact us to find out more.

About

VictoriaMetrics Anomaly Detection is a service that continuously scans Victoria Metrics time series and detects unexpected changes within data patterns in real-time. It does so by utilizing user-configurable machine learning models.

It periodically queries user-specified metrics, computes an “anomaly score” for them, based on how well they fit a predicted distribution, taking into account periodical data patterns with trends, and pushes back the computed “anomaly score” to Victoria Metrics. Then, users can enable alerting rules based on the “anomaly score”.

Compared to classical alerting rules, anomaly detection is more “hands-off” i.e. it allows users to avoid setting up manual alerting rules set up and catching anomalies that were not expected to happen. In other words, by setting up alerting rules, a user must know what to look for, ahead of time, while anomaly detection looks for any deviations from past behavior.

In addition to that, setting up alerting rules manually has been proven to be tedious and error-prone, while anomaly detection can be easier to set up, and use the same model for different metrics.

How?

Victoria Metrics Anomaly Detection service (vmanomaly) allows you to apply several built-in anomaly detection algorithms. You can also plug in your own detection models, code doesn’t make any distinction between built-in models or external ones.

All the service parameters (model, schedule, input-output) are defined in a config file.

Single config file supports only one model, but it’s totally OK to run multiple vmanomaly processes in parallel, each using its own config.

Models

Currently, vmanomaly ships with a few common models:

1. ZScore

   (useful for testing)

   Simplistic model, that detects outliers as all the points that lie farther than a certain amount from time-series mean (straight line). Keeps only two model parameters internally: mean and std (standard deviation).

2. Prophet

   (simplest in configuration, recommended for getting starting)

   Uses Facebook Prophet for forecasting. The anomaly score is computed of how close the actual time series values follow the forecasted values (yhat), and whether it’s within forecasted bounds (yhat_lower, yhat_upper). The anomaly score reaches 1.0 if the actual data values are equal to yhat_lower or yhat_upper. The anomaly score is above 1.0 if the actual data values are outside the yhat_lower/yhat_upper bounds.

   See Prophet documentation

3. Holt-Winters

   Very popular forecasting algorithm. See statsmodels.org documentation for Holt-Winters exponential smoothing.

4. Seasonal-Trend Decomposition

   Extracts three components: season, trend, and residual, that can be plotted individually for easier debugging. Uses LOESS (locally estimated scatterplot smoothing). See statsmodels.org documentation for LOESS STD.

5. ARIMA

   Commonly used forecasting model. See statsmodels.org documentation for ARIMA.

6. Rolling Quantile

   A simple moving window of quantiles. Easy to use, easy to understand, but not as powerful as other models.

Examples

For example, here’s how Prophet predictions could look like on a real-data example
(Prophet auto-detected seasonality interval):

And here’s what Holt-Winters predictions real-world data could look like (seasonality manually set to 1 week). Notice that it predicts anomalies in different places than Prophet because the model noticed there are usually spikes on Friday morning, so it accounted for that:

Process

Upon starting, vmanomaly queries the initial range of data, and trains its model (“fit” by convention).

Then, reads new data from VictoriaMetrics, according to schedule, and invokes its model to compute “anomaly score” for each data point. The anomaly score ranges from 0 to positive infinity. Values less than 1.0 are considered “not an anomaly”, values greater or equal than 1.0 are considered “anomalous”, with greater values corresponding to larger anomaly. Then, VMAnomaly pushes the metric to vminsert (under the user-configured metric name, optionally preserving labels).

Usage

The vmanomaly accepts only one parameter -- config file path:

python3 vmanomaly.py config_zscore.yaml

or

python3 -m vmanomaly config_zscore.yaml

It is also possible to split up config into multiple files, just list them all in the command line:

python3 -m vmanomaly model_prophet.yaml io_csv.yaml scheduler_oneoff.yaml