Basic terms in Monium
This section describes the terms used in Monium and its guides.
Common terms in Monium
Quick links
Quick links are links displayed in the Monium interface that you can configure for each project via the UI or API.
Timeseries
Timeseries is a time-ordered set of values of a variable with a timestamp for each measurement: [[(time_stamp_1, value_1), (time_stamp_2, value_2), ...]. For more information, see Data model in Monium.Metrics.
Chart
Dashboard
Dasboard is a Monium configuration object that includes sets of charts and their display settings.
Cluster
Cluster is a Monium configuration object defining the set of hosts Monium will poll to collect metrics in metric Pull mode. The cluster is usually aligned with the environment (installation) of the client application delivering the metrics, e.g., production, test, or regional clusters.
Label
Label is a pair of strings in <name>="<value>" format, e.g., host="monitoring-kfront-man-00" or code="200". Labels describe various slices used to visualize metrics delivered to Monium. A set of labels identify a metric. For more information, see Data model in Monium.Metrics.
Metric
Metric is a timeseries and its identifying set of labels enclosed in curly brackets.
Dashboard parameter
Dashboard parameters let you create interactive dashboards whose content changes depending on the user's choice. For example, a dashboard presenting aggregated VM status information can be parameterized using the VM ID.
Project
Project is a top-level Monium configuration object containing observation data. It is used for logical grouping of related services and assignment of access permissions for telemetry data to project teams. A project can contain data on resources of an individual folder, in which case the project ID format is folder_<folder_ID>. With a project, you can restrict access to data by assigning roles.
Decimation
-
Read data decimation is a mechanism that limits the number of data points on the chart. For more information, see Read data decimation.
-
Historical data decimation is a mechanism used to archive historical data in Monium. To reduce the amount of stored historical data, all metrics saved in Monium get aggregated on a 5-minute grid after a week. For more information, see Historical data decimation.
Service
Service is a Monium configuration object that defines the methods used to deliver data to Monium (Pull or Push). The service is usually aligned with the client application delivering the metrics, e.g., a web server, database, or another component.
Shard
Shard is a Monium configuration object that owns the metrics from a particular service-cluster pair, as well as their storage settings, e.g., TTL.
Expression Language
Expression Language (or simply Expression) is a method to describe how to calculate and transform metrics when configuring charts and alerts and exporting data over the API. For more information, see Query language in Monium.Metrics.
TTL (Time-to-Live)
TTL is a mechanism used to automatically delete obsolete metrics that have not been receiving new values for a given period of time.
For example, a TTL of 7 means that metrics whose last point's timestamp is more than 7 days in the past will be automatically deleted.
For more information, see TTL (Time-to-Live)
Unified Agent
Unified Agent is a program used to collect, store, transform, and deliver metrics and logs to Monium.
Logging terms
Log line
A log line is the smallest data processing unit in the system. It represents a system event, its description, its timestamp in the user’s system, and labels that identify it.
Log line size
The log line size is measured in bytes and equals the sum of all top-level field values, as well as the sum of lengths of all keys and values from labels and meta.
Event lag
Event lag is the latency between when the event was generated in the user system and when it was delivered to the centralized logging system (upstream of the collector).
Observed lag
Observed lag is the latency between when the event was first processed by the agent (the very first one in the chain, if any) and when it was delivered to the centralized logging system (upstream of the collector).
Tracing terms
Request tracing
Request tracing is there for analysis of distributed requests. It allows you to visualize and track the execution path of a specific request as it passes through many different services and infrastructure components. Request execution path is a sequence of operations called spans.
Span
A span is a basic distributed tracing element representing a single operation in your system. For example, this may be a database query, an HTTP request, or a function call. Each span comes with a name, start and end time, labels, logs, and execution context. Spans may contain links to other spans to join them together into a trace.
Trace
A trace is a combination of spans forming the execution path of a particular request.