Renders a table using ASCII characters.
Renders a table using ASCII characters. The table is represented by a header row, consisting of ASCIITableHeader objects which define the column label and the column alignment, and multiple rows, each of which contains an array of values to go in each row.
There must be an equal number of items in each row and in the header.
to be removed in version 0.3.0
to be removed in version 0.3.0
to be removed in version 0.3.0
to be removed in version 0.3.0
Allows metrics to be created for an application.
Allows metrics to be created for an application. Currently only timers are supported, but other types of metrics may be supported in the future.
Classes should extend this class to provide collections of metrics for an application. For example, typical usage might be:
object Timers extends Metrics { val Operation1 = timer("Operation 1") val Operation2 = timer("Operation 2") }
This creates two timers: one for Operation 1, and one for Operation 2.
To switch-on recording of metrics for a specific thread, the companion Metrics object must be used.
to be removed in version 0.3.0
Spark listener that accumulates metrics in the passed-in RecordedMetrics object at stage completion time.
Spark listener that accumulates metrics in the passed-in RecordedMetrics object at stage completion time.
to be removed in version 0.3.0
This class relies on being run in the same process as the driver. However, this is the way that Spark seems to work.
Allows metrics to be recorded.
Allows metrics to be recorded. Currently only timings are supported, but other metrics
may be supported in the future. Use the startPhase
method to start recording a timing
and the finishPhase
method to finish recording it. If timings are nested the
hierarchy is preserved.
Note: this class is intended to be used in a thread-local context, and therefore it is not thread-safe. Do not attempt to call it concurrently from multiple threads!
to be removed in version 0.3.0
Tabular representation of a set of Monitors.
Tabular representation of a set of Monitors. A MonitorTable consists of a list of TableHeaders, which specify not only the titles of the columns, but also how the data is extracted from the rows, as well as a list of rows which are Monitors containing the data. Each header can specify the following:
- name: the name of the column; this will be the text that is used in the header row - valueExtractor: specifies how data will be extracted from a monitor; the functions in the ValueExtractor object can be used to create ValueExtractors for a particular purpose - formatFunction (optional): a function which formats the value extracted from the monitor as a String - alignment (optional): specifies how the data in a cell is aligned; defaults to right-alignment
to be removed in version 0.3.0
Currently this consists only of metrics collected from Spark, but there may be others in the future.
Currently this consists only of metrics collected from Spark, but there may be others in the future.
to be removed in version 0.3.0
Specifies a timing that is to recorded
Specifies a timing that is to recorded
to be removed in version 0.3.0
Timer that implements the com.netflix.servo.monitor.CompositeMonitor interface.
Timer that implements the com.netflix.servo.monitor.CompositeMonitor interface. Sub-monitors contain specific metrics, and can be identified using the following tags: - Total Time: ServoTimer.TotalTimeTag - Count: ServoTimer.CountTag - Mean: ServoTimer.MeanTag - Max: ServoTimer.MaxTag - Min: ServoTimer.MinTag All timings are returned in nanoseconds.
to be removed in version 0.3.0
Holds a collection of ServoTimers.
Holds a collection of ServoTimers. Each instance of a timer is stored against a TimingPath, which
specifies all of its ancestors. Timings can be recorded using the recordTiming
method, which will
either update an existing timer if the specified TimingPath exists already, or will create a new timer.
to be removed in version 0.3.0
Implementation of AccumulableParam that records timings and returns a ServoTimers with the accumulated timings.
Implementation of AccumulableParam that records timings and returns a ServoTimers with the accumulated timings.
to be removed in version 0.3.0
Allows metrics for Spark to be captured and rendered in tabular form.
Allows metrics for Spark to be captured and rendered in tabular form.
to be removed in version 0.3.0
to be removed in version 0.3.0
to be removed in version 0.3.0
Specifies the title of a column, as well as how data is extracted to form the cells of this column
Specifies the title of a column, as well as how data is extracted to form the cells of this column
to be removed in version 0.3.0
to be removed in version 0.3.0
to be removed in version 0.3.0
Specifies a timer name, along with all of its ancestors.
Specifies a timer name, along with all of its ancestors.
to be removed in version 0.3.0
to be removed in version 0.3.0
to be removed in version 0.3.0
Represents alignment of a cell in a table
Represents alignment of a cell in a table
to be removed in version 0.3.0
Functions for formatting durations.
Functions for formatting durations. The following rules are applied: - Durations greater than one hour are formatted as "X hours X mins X secs" - Durations greater than one minute, but less than one hour are formatted as "X mins X secs" - Durations less than one minute are formatted with two digits after the decimal point. Zeros are suppressed.
to be removed in version 0.3.0
Helper functions for instrumentation
Helper functions for instrumentation
to be removed in version 0.3.0
Manages recording of metrics in an application.
Manages recording of metrics in an application. Metrics collection is turned on for a particular thread by
calling the initialize
method. After initialize
is called the Metrics.Recorder field is set, and the
application can use that to record metrics. To print metrics, call the print
method, and to stop recording metrics
for a particular thread, call the stopRecording
method.
If an application does not want to record metrics, it can simply avoid calling the initialize
method (or can
call the stopRecording
method if there is a chance that initialize
has been called on that thread previously).
Attempting to record metrics when the initialize method has not been called will not produce an error, and incurs very
little overhead. However, attempting to call the print method to print the metrics will produce an error.
to be removed in version 0.3.0
to be removed in version 0.3.0
to be removed in version 0.3.0
to be removed in version 0.3.0
Deprecated in version 0.2.16, will be removed in version 0.3.0.