Skip to main content

Metrics Quantization

Sumo ingests individual metric data points from your metric sources. In metric visualizations, rather than charting individual data points, Sumo presents the aggregated value of the data points received during an interval.

Quantization is the process of aggregating metric data points for time series over an interval, for example, an hour or a minute, using a particular aggregation function: avg, min, max, sum, or count.

Quantization terminology​

This section defines the quantization-related terms we use in Sumo. So, what is quantization? At a high level, it’s the process that Sumo Logic performs on raw data points to produce the aggregated metric values that your metric queries run against.

Buckets​

We use the term bucket to refer to the intervals across which Sumo quantizes your metrics.

When you run a metric query, Sumo divides up your metric query time range into contiguous buckets, either automatically, or based on the interval you specify in the quantize operator. For example, given this query:

metric=CPU_Idle | quantize to 15m

Sumo divides your time range into 15 minute buckets.

For each bucket, Sumo uses a rollup type, described below, to aggregate the values of all the data points in the bucket. The aggregated values are displayed in your metric visualization or processed further in the pipeline.

By default, Sumo uses the avg rollup type. You can specify another rollup type by using the quantize operator, as described in Quantize with rollup type specified below.

Rollup types​

We use the term rollup to refer to the aggregation function Sumo uses when quantizing metrics. This table describes the different rollup types you can select when running a query.

Rollup type Description
avg Calculates the average value of the data points for a time series in each bucket.
min Calculates the minimum value among the data points for a time series in each bucket.
max Calculates the maximum value among the data points for a time series in each bucket.
sumCalculates the sum of the values of the data points for a time series in each bucket.
countCalculates the count of data points for a time series in each bucket.

Sumo quantizes metrics upon ingestion and at query time.

Quantization at ingestion​

Upon ingestion, Sumo quantizes raw metric data points to one hour resolutions for all rollup types: avg, min, max, sum, and count. This data is stored in one hour rollup tables in Sumo. The raw data is stored in a table referred to as the baseline table. For information about retention times, see Metric Ingestion and Storage.

Automatic quantization at query time​

This section describes how Sumo quantizes metrics when you run a metric query without specifying quantization interval.

If you do not use the quantize to INTERVAL in your metric query, Sumo automatically determines an optimal quantization interval, based on the age of the data you are querying and the query time range. The quantization interval is shown at the top of the metric query tab.

metrics

The age of the metrics in the time range governs the minimum quantization interval (based on what rollups are available for the query time range). Sumo retains only the last 30 days of raw metric data. So, when you query metrics that are more than 30 days old, Sumo must quantize the data to at least 1 hour, because that’s the minimum resolution rollup available given the age of the data.

If you want, you can override the automatic quantization interval. In the Metrics Explorer’s basic mode you can set the quantization interval in the row creator in the UI. In advanced mode, use the quantize operator and specify the interval that fits your need

Sumo Logic sets the actual quantization interval to be as close to your selection as possible. If it is not possible to set the actual interval to the targeted interval—typically because too many buckets would be produced to reasonably show on the chart—Sumo displays a message like the following:

metrics

Sumo Logic will never decrease the quantization interval that you specify. We’ll either use that interval, or increase it as appropriate.

How Sumo chooses rollup table and quantization interval​

If you do not specify a rollup type in your query, Sumo Logic will run the query using the avg rollup.

The table below shows how Sumo Logic selects a quantization interval based on query time range, in the case that you do not specify those options explicitly using the quantize operator.

Query time rangeDefault quantization interval
400 days1 day
200 days1 day
150 days12 hours
90 days6 hours
30 days2 hours
14 days1 hour
7 days1 hour
3 days1 hour
2 days10 minutes
1 day5 minutes
6 hours1 minute
3 hours30 seconds
1 hour15 seconds

Explicit quantization at query time​

When you run a metric query, you can optionally use Metrics Quantize Operator to specify a quantization interval, rollup type, or both.

When you run a query with the quantize operator, the way that Sumo quantizes your metric data points depends on the rollup type you specify, if any, in the quantize clause of your query. Rollup types include avg, min, max, sum, and count.

note

Specifying rollup type and quantization interval is optional for the quantize operator, however, one of them needs to be present.

Quantize with rollup type specified​

To specify the rollup type for quantization, include the quantize operator as the first operator in your query (immediately after the selector), and specify the rolloup type with the using clause. For example, given this query:

metric=CPU_Idle | quantize to 15m using sum

Sumo will quantize to the sum rollup type.

Also, you can skip specifying target quantization interval:

metric=CPU_Idle | quantize using max

In this case, Sumo Logic will automatically determine appropriate quantization interval and quantize to the max rollup type.

note

If the quantize operator in your query is preceded by another metrics operator, a rollup type you specify with using will be ignored – it will not be applied at the selector level.

Quantize with no rollup type specified​

If your metrics query uses the quantize operator without specifying a rollup type, internally, Sumo Logic uses the default avg rollup.

For example, given this query:

metric=CPU_Idle | quantize to 15m

15-minute interval and avg rollup will be used.

Status
Legal
Privacy Statement
Terms of Use

Copyright © 2024 by Sumo Logic, Inc.