5/2/2023 0 Comments Graphite timeslice example![]() Similarly, the second row computes the sum of all metrics with name “req_total”.įinally the “/” operator, represented by the function “op:div2()”, divides the first sum by the second sum to get ratio of failed requests. The resulting streams are “piped” into the stats:sum() function, which sums data from all input streams. In the first row, the find("req_error") function selects data from all metrics with name “req_error”. This statement gives rise to the following data-flow graph: To get an idea of the syntax, consider the following CAQL statement: ( find("req_error") | stats:sum() ) / The precise semantics of CAQL functions and streams are will be explained in the following paragraphs. Unlike metrics, CAQL streams are not persisted and have a simplified data model.ĬAQL checks can be used to create a metric from a CAQL stream.Ī CAQL statement consists of one or more functions that can be composed in different ways.įunctions can have multiple input and output streams. A set of tags, that is used by functions like group_byĮach Circonus metric gives rise to a stream of values.A label, which is used for the graph legend.If no data was collected for a given period, then a NULL value representing missing data is inserted. Streams are synchronous, in the sense that for each sampling period of 1m, there is precisely one value. histogram values, consisting of histograms.numeric values, double precision floating point numbers.CAQL Data Flow and Data StructuresĬAQL is a stream processing language that allows the user to create complex transformations of metric data.Ī CAQL stream is time-indexed list of values, that can have one of the following types: This manual contains formal descriptions of the CAQL syntax and semantics.įor a tutorial, please refer to the CAQL Getting Started page.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |