improved scheduling for urgent tasks, using farm of replacement threads (may lead to factor 2 overloading, but CPUs are usually hyperthreaded);
/* Title: Pure/Tools/ml_statistics.scala
Author: Makarius
ML runtime statistics.
*/
package isabelle
import scala.collection.mutable
import scala.collection.immutable.{SortedSet, SortedMap}
import scala.swing.{Frame, Component}
import org.jfree.data.xy.{XYSeries, XYSeriesCollection}
import org.jfree.chart.{JFreeChart, ChartPanel, ChartFactory}
import org.jfree.chart.plot.PlotOrientation
object ML_Statistics
{
/* content interpretation */
final case class Entry(time: Double, data: Map[String, Double])
def apply(name: String, stats: List[Properties.T]): ML_Statistics =
new ML_Statistics(name, stats)
def apply(info: Build.Log_Info): ML_Statistics =
apply(info.name, info.stats)
val empty = apply("", Nil)
/* standard fields */
val tasks_fields =
("Future tasks",
List("tasks_ready", "tasks_pending", "tasks_running", "tasks_passive", "tasks_urgent"))
val workers_fields =
("Worker threads", List("workers_total", "workers_active", "workers_waiting"))
val GC_fields = ("GCs", List("partial_GCs", "full_GCs"))
val heap_fields =
("Heap", List("size_heap", "size_allocation", "size_allocation_free",
"size_heap_free_last_full_GC", "size_heap_free_last_GC"))
val threads_fields =
("Threads", List("threads_total", "threads_in_ML", "threads_wait_condvar",
"threads_wait_IO", "threads_wait_mutex", "threads_wait_signal"))
val time_fields = ("Time", List("time_CPU", "time_GC"))
val speed_fields = ("Speed", List("speed_CPU", "speed_GC"))
val standard_fields =
List(tasks_fields, workers_fields, GC_fields, heap_fields, threads_fields,
time_fields, speed_fields)
}
final class ML_Statistics private(val name: String, val stats: List[Properties.T])
{
val Now = new Properties.Double("now")
def now(props: Properties.T): Double = Now.unapply(props).get
require(stats.forall(props => Now.unapply(props).isDefined))
val time_start = if (stats.isEmpty) 0.0 else now(stats.head)
val duration = if (stats.isEmpty) 0.0 else now(stats.last) - time_start
val fields: Set[String] =
SortedSet.empty[String] ++
(for (props <- stats.iterator; (x, _) <- props.iterator if x != Now.name)
yield x)
val content: List[ML_Statistics.Entry] =
{
var last_edge = Map.empty[String, (Double, Double, Double)]
val result = new mutable.ListBuffer[ML_Statistics.Entry]
for (props <- stats) {
val time = now(props) - time_start
require(time >= 0.0)
// rising edges -- relative speed
val speeds =
for ((key, value) <- props; a <- Library.try_unprefix("time", key)) yield {
val (x0, y0, s0) = last_edge.getOrElse(a, (0.0, 0.0, 0.0))
val x1 = time
val y1 = java.lang.Double.parseDouble(value)
val s1 = if (x1 == x0) 0.0 else (y1 - y0) / (x1 - x0)
val b = ("speed" + a).intern
if (y1 > y0) { last_edge += (a -> (x1, y1, s1)); (b, s1) } else (b, s0)
}
val data =
SortedMap.empty[String, Double] ++ speeds ++
(for ((x, y) <- props.iterator if x != Now.name)
yield (x, java.lang.Double.parseDouble(y)))
result += ML_Statistics.Entry(time, data)
}
result.toList
}
/* charts */
def update_data(data: XYSeriesCollection, selected_fields: Iterable[String])
{
data.removeAllSeries
for {
field <- selected_fields.iterator
series = new XYSeries(field)
} {
content.foreach(entry => series.add(entry.time, entry.data(field)))
data.addSeries(series)
}
}
def chart(title: String, selected_fields: Iterable[String]): JFreeChart =
{
val data = new XYSeriesCollection
update_data(data, selected_fields)
ChartFactory.createXYLineChart(title, "time", "value", data,
PlotOrientation.VERTICAL, true, true, true)
}
def chart(arg: (String, Iterable[String])): JFreeChart = chart(arg._1, arg._2)
def show_standard_frames(): Unit =
ML_Statistics.standard_fields.map(chart(_)).foreach(c =>
GUI_Thread.later {
new Frame {
iconImage = GUI.isabelle_image()
title = name
contents = Component.wrap(new ChartPanel(c))
visible = true
}
})
}