FuseQL is a new query language to perform analytics on log events. It is meant to be a drop-in replacement for Grafana’s logQL support in Kloudfuse. In addition, it also supports more advanced operators such as anomaly detection, outlier detection, forecasting and several arithmetic and trigonometric operators.
All the various operators supported in FuseQL are documented in the following pages:
Users can switch between FuseQL and logQL with a toggle button on the log analytics page.
Query Builder
Log Filters
To filter out logs that you want to run your analytics query on, click 
(Search logs). If no filters are specified, FuseQL will consider all log lines.
Aggregations
Once the log filters have been added, you can apply an aggregation operator on the log lines. By default, FuseQL will display the count of all log lines that match the filter conditions.
For more information on various supported aggregation operators, refer to Aggregation Operators.
Group Bys
FuseQL also supports grouping of log lines by label and facets. Click on the by dropdown to add group by fields to your aggregation operators. By default, FuseQL groups by everything, collapsing all the columns.
Limit To
By default, FuseQL limits the resulting time series values to the top 10. This can be changed by editing the dropdown.
Roll Up
FuseQL aggregates data by roll up time period (for example, five seconds). The roll up value is auto-selected by the user-selected time range, but can be edited if needed. Editing the roll up will change the number of bucketed results you see in the visualization.
Below is an example of using a 1m roll up in a time range of 5m resulting in 5 buckets.
Add Query
To add a query, click 
(Add Query). Notice that a new query 
(B) appears under query 
(A), and it is a duplicate.
Add Formula
Visualizations
FuseQL supports four different visualization types. The default visualization is Time Series.
Time Series
Use Time Series Visualization when you want to analyze trends, patterns, and behaviors over time.
Top List
Table
Pie Chart
Algorithms
Overlay a band on the metric, showing the expected behavior of a series based on past values.
Highlight outliers series.
Forecast future values based on past values.