Fix benchmark results script

benchmarks/README.md

@@ -1,4 +1,119 @@
-# Module benchmarks
+# BenchmarksResult
+
+## Report for benchmark configuration <code>main</code>
+
+* Run on Java HotSpot(TM) 64-Bit Server VM (build 21.0.4+8-LTS-jvmci-23.1-b41) with Java process:
+
+```
+C:\Users\altavir\scoop\apps\graalvm-oracle-21jdk\current\bin\java.exe -XX:ThreadPriorityPolicy=1 -XX:+UnlockExperimentalVMOptions -XX:+EnableJVMCIProduct -XX:-UnlockExperimentalVMOptions -Dfile.encoding=UTF-8 -Duser.country=US -Duser.language=en -Duser.variant
+```
+* JMH 1.21 was used in `thrpt` mode with 5 warmup iterations by 10 s and 5 measurement iterations by 10 s.
+### [ArrayBenchmark](src/jvmMain/kotlin/space/kscience/kmath/benchmarks/ArrayBenchmark.kt)
+
+| Benchmark | Score |
+|:---------:|:-----:|
+|`benchmarkArrayRead`|1.9E+07 &plusmn; 2.3E+05 ops/s|
+|`benchmarkBufferRead`|1.4E+07 &plusmn; 8.7E+05 ops/s|
+|`nativeBufferRead`|1.4E+07 &plusmn; 1.3E+06 ops/s|
+### [BigIntBenchmark](src/jvmMain/kotlin/space/kscience/kmath/benchmarks/BigIntBenchmark.kt)
+
+| Benchmark | Score |
+|:---------:|:-----:|
+|`jvmAdd`|5.1E+07 &plusmn; 1.3E+06 ops/s|
+|`jvmAddLarge`|5.1E+04 &plusmn; 8.2E+02 ops/s|
+|`jvmMultiply`|8.5E+07 &plusmn; 9.7E+06 ops/s|
+|`jvmMultiplyLarge`|2.5E+02 &plusmn; 15 ops/s|
+|`jvmParsing10`|8.7E+06 &plusmn; 5.1E+05 ops/s|
+|`jvmParsing16`|6.4E+06 &plusmn; 1.8E+05 ops/s|
+|`jvmPower`|28 &plusmn; 0.79 ops/s|
+|`jvmSmallAdd`|7.0E+07 &plusmn; 4.3E+06 ops/s|
+|`kmAdd`|4.8E+07 &plusmn; 2.2E+06 ops/s|
+|`kmAddLarge`|3.5E+04 &plusmn; 3.7E+03 ops/s|
+|`kmMultiply`|6.7E+07 &plusmn; 1.5E+07 ops/s|
+|`kmMultiplyLarge`|54 &plusmn; 4.2 ops/s|
+|`kmParsing10`|4.5E+06 &plusmn; 8.3E+04 ops/s|
+|`kmParsing16`|4.9E+06 &plusmn; 1.1E+05 ops/s|
+|`kmPower`|10 &plusmn; 0.96 ops/s|
+|`kmSmallAdd`|4.1E+07 &plusmn; 5.9E+05 ops/s|
+### [BufferBenchmark](src/jvmMain/kotlin/space/kscience/kmath/benchmarks/BufferBenchmark.kt)
+
+| Benchmark | Score |
+|:---------:|:-----:|
+|`bufferViewReadWrite`|5.8E+06 &plusmn; 1.6E+05 ops/s|
+|`bufferViewReadWriteSpecialized`|5.6E+06 &plusmn; 2.6E+05 ops/s|
+|`complexBufferReadWrite`|6.6E+06 &plusmn; 2.7E+05 ops/s|
+|`doubleArrayReadWrite`|7.5E+06 &plusmn; 1.0E+06 ops/s|
+|`doubleBufferReadWrite`|8.0E+06 &plusmn; 6.7E+05 ops/s|
+### [DotBenchmark](src/jvmMain/kotlin/space/kscience/kmath/benchmarks/DotBenchmark.kt)
+
+| Benchmark | Score |
+|:---------:|:-----:|
+|`bufferedDot`|1.3 &plusmn; 0.020 ops/s|
+|`cmDot`|0.47 &plusmn; 0.42 ops/s|
+|`cmDotWithConversion`|0.76 &plusmn; 0.13 ops/s|
+|`ejmlDot`|6.7 &plusmn; 0.091 ops/s|
+|`ejmlDotWithConversion`|6.4 &plusmn; 0.82 ops/s|
+|`multikDot`|40 &plusmn; 6.7 ops/s|
+|`parallelDot`|12 &plusmn; 1.8 ops/s|
+|`tensorDot`|1.2 &plusmn; 0.041 ops/s|
+|`tfDot`|5.9 &plusmn; 0.49 ops/s|
+### [ExpressionsInterpretersBenchmark](src/jvmMain/kotlin/space/kscience/kmath/benchmarks/ExpressionsInterpretersBenchmark.kt)
+
+| Benchmark | Score |
+|:---------:|:-----:|
+|`asmGenericExpression`|29 &plusmn; 1.2 ops/s|
+|`asmPrimitiveExpression`|43 &plusmn; 1.3 ops/s|
+|`asmPrimitiveExpressionArray`|71 &plusmn; 0.38 ops/s|
+|`functionalExpression`|5.6 &plusmn; 0.11 ops/s|
+|`justCalculate`|69 &plusmn; 9.0 ops/s|
+|`mstExpression`|7.1 &plusmn; 0.020 ops/s|
+|`rawExpression`|41 &plusmn; 1.5 ops/s|
+### [IntegrationBenchmark](src/jvmMain/kotlin/space/kscience/kmath/benchmarks/IntegrationBenchmark.kt)
+
+| Benchmark | Score |
+|:---------:|:-----:|
+|`complexIntegration`|3.6E+03 &plusmn; 1.9E+02 ops/s|
+|`doubleIntegration`|3.7E+03 &plusmn; 12 ops/s|
+### [JafamaBenchmark](src/jvmMain/kotlin/space/kscience/kmath/benchmarks/JafamaBenchmark.kt)
+
+| Benchmark | Score |
+|:---------:|:-----:|
+|`core`|38 &plusmn; 0.64 ops/s|
+|`jafama`|52 &plusmn; 0.36 ops/s|
+|`strictJafama`|52 &plusmn; 4.0 ops/s|
+### [MatrixInverseBenchmark](src/jvmMain/kotlin/space/kscience/kmath/benchmarks/MatrixInverseBenchmark.kt)
+
+| Benchmark | Score |
+|:---------:|:-----:|
+|`cmLUPInversion`|2.2E+03 &plusmn; 76 ops/s|
+|`ejmlInverse`|1.3E+03 &plusmn; 5.7 ops/s|
+|`kmathLupInversion`|9.5E+02 &plusmn; 1.8E+02 ops/s|
+|`kmathParallelLupInversion`|9.1E+02 &plusmn; 1.4E+02 ops/s|
+### [NDFieldBenchmark](src/jvmMain/kotlin/space/kscience/kmath/benchmarks/NDFieldBenchmark.kt)
+
+| Benchmark | Score |
+|:---------:|:-----:|
+|`boxingFieldAdd`|7.7 &plusmn; 0.79 ops/s|
+|`multikAdd`|6.5 &plusmn; 0.33 ops/s|
+|`multikInPlaceAdd`|64 &plusmn; 0.79 ops/s|
+|`specializedFieldAdd`|8.0 &plusmn; 0.090 ops/s|
+|`tensorAdd`|9.2 &plusmn; 0.053 ops/s|
+|`tensorInPlaceAdd`|17 &plusmn; 10 ops/s|
+|`viktorAdd`|7.6 &plusmn; 1.2 ops/s|
+### [ViktorBenchmark](src/jvmMain/kotlin/space/kscience/kmath/benchmarks/ViktorBenchmark.kt)
+
+| Benchmark | Score |
+|:---------:|:-----:|
+|`doubleFieldAddition`|7.7 &plusmn; 0.34 ops/s|
+|`rawViktor`|5.9 &plusmn; 1.1 ops/s|
+|`viktorFieldAddition`|7.3 &plusmn; 1.1 ops/s|
+### [ViktorLogBenchmark](src/jvmMain/kotlin/space/kscience/kmath/benchmarks/ViktorLogBenchmark.kt)
+
+| Benchmark | Score |
+|:---------:|:-----:|
+|`rawViktorLog`|1.4 &plusmn; 0.076 ops/s|
+|`realFieldLog`|1.3 &plusmn; 0.069 ops/s|
+|`viktorFieldLog`|1.3 &plusmn; 0.032 ops/s|
 
 
 

benchmarks/build.gradle.kts

@@ -3,10 +3,6 @@ import com.fasterxml.jackson.module.kotlin.readValue
 import kotlinx.benchmark.gradle.BenchmarksExtension
 import java.time.LocalDateTime
 import java.time.ZoneId
-import java.time.format.DateTimeFormatter
-import java.time.format.DateTimeFormatterBuilder
-import java.time.format.SignStyle
-import java.time.temporal.ChronoField.*
 import java.util.*
 
 plugins {
@@ -220,41 +216,20 @@ readme {
 
     val jsonMapper = jacksonObjectMapper()
 
-
-    val ISO_DATE_TIME: DateTimeFormatter = DateTimeFormatterBuilder().run {
-        parseCaseInsensitive()
-        appendValue(YEAR, 4, 10, SignStyle.EXCEEDS_PAD)
-        appendLiteral('-')
-        appendValue(MONTH_OF_YEAR, 2)
-        appendLiteral('-')
-        appendValue(DAY_OF_MONTH, 2)
-        appendLiteral('T')
-        appendValue(HOUR_OF_DAY, 2)
-        appendLiteral('.')
-        appendValue(MINUTE_OF_HOUR, 2)
-        optionalStart()
-        appendLiteral('.')
-        appendValue(SECOND_OF_MINUTE, 2)
-        optionalStart()
-        appendFraction(NANO_OF_SECOND, 0, 9, true)
-        optionalStart()
-        appendOffsetId()
-        optionalStart()
-        appendLiteral('[')
-        parseCaseSensitive()
-        appendZoneRegionId()
-        appendLiteral(']')
-        toFormatter()
-    }
-
     fun noun(number: Number, singular: String, plural: String) = if (number.toLong() == 1L) singular else plural
 
     extensions.findByType(BenchmarksExtension::class.java)?.configurations?.forEach { cfg ->
-        property("benchmark${cfg.name.replaceFirstChar { if (it.isLowerCase()) it.titlecase(Locale.getDefault()) else it.toString() }}") {
+        val propertyName =
-            val launches = layout.buildDirectory.dir("reports/benchmarks/${cfg.name}").get()
+            "benchmark${cfg.name.replaceFirstChar { if (it.isLowerCase()) it.titlecase(Locale.getDefault()) else it.toString() }}"
 
-            val resDirectory = launches.files().maxByOrNull {
+        logger.info("Processing benchmark data from benchmark ${cfg.name} into readme property $propertyName")
-                LocalDateTime.parse(it.name, ISO_DATE_TIME).atZone(ZoneId.systemDefault()).toInstant()
+
+        val launches = layout.buildDirectory.dir("reports/benchmarks/${cfg.name}").get().asFile
+        if (!launches.exists()) return@forEach
+
+        property(propertyName) {
+            val resDirectory = launches.listFiles()?.maxByOrNull {
+                LocalDateTime.parse(it.name).atZone(ZoneId.systemDefault()).toInstant()
             }
 
             if (resDirectory == null || !(resDirectory.resolve("jvm.json")).exists()) {
@@ -264,10 +239,7 @@ readme {
                     jsonMapper.readValue<List<JmhReport>>(resDirectory.resolve("jvm.json"))
 
                 buildString {
-                    appendLine("<details>")
+                    appendLine("## Report for benchmark configuration <code>${cfg.name}</code>")
-                    appendLine("<summary>")
-                    appendLine("Report for benchmark configuration <code>${cfg.name}</code>")
-                    appendLine("</summary>")
                     appendLine()
                     val first = reports.first()
 
@@ -289,15 +261,19 @@ readme {
                         } by ${first.measurementTime}."
                     )
 
+                    reports.groupBy { it.benchmark.substringBeforeLast(".") }.forEach { (cl, compare) ->
+                        appendLine("### [${cl.substringAfterLast(".")}](src/jvmMain/kotlin/${cl.replace(".","/")}.kt)")
                         appendLine()
                         appendLine("| Benchmark | Score |")
                         appendLine("|:---------:|:-----:|")
+                        compare.forEach { report ->
+                            val benchmarkName = report.benchmark.substringAfterLast(".")
+                            val score = String.format("%.2G", report.primaryMetric.score)
+                            val error = String.format("%.2G", report.primaryMetric.scoreError)
 
-                    reports.forEach { report ->
+                            appendLine("|`$benchmarkName`|$score &plusmn; $error ${report.primaryMetric.scoreUnit}|")
-                        appendLine("|`${report.benchmark}`|${report.primaryMetric.score} &plusmn; ${report.primaryMetric.scoreError} ${report.primaryMetric.scoreUnit}|")
+                        }
-                    }
+                    }
-
-                    appendLine("</details>")
                 }
             }
         }

benchmarks/docs/README-TEMPLATE.md

@@ -0,0 +1,5 @@
+# BenchmarksResult
+
+${benchmarkMain}
+
+

examples/src/main/kotlin/space/kscience/kmath/stat/DistributionBenchmark.kt

@@ -36,6 +36,26 @@ private suspend fun runKMathChained(): Duration {
     return Duration.between(startTime, Instant.now())
 }
 
+private fun runKMathBlocking(): Duration {
+    val generator = RandomGenerator.fromSource(RandomSource.MT, 123L)
+    val normal = GaussianSampler(7.0, 2.0)
+    val chain = normal.sample(generator)
+    val startTime = Instant.now()
+    var sum = 0.0
+
+    repeat(10000001) { counter ->
+        sum += chain.nextBlocking()
+
+        if (counter % 100000 == 0) {
+            val duration = Duration.between(startTime, Instant.now())
+            val meanValue = sum / counter
+            println("Chain sampler completed $counter elements in $duration: $meanValue")
+        }
+    }
+
+    return Duration.between(startTime, Instant.now())
+}
+
 private fun runCMDirect(): Duration {
     val rng = RandomSource.MT.create(123L)
 
@@ -67,6 +87,8 @@ private fun runCMDirect(): Duration {
 fun main(): Unit = runBlocking(Dispatchers.Default) {
     val directJob = async { runCMDirect() }
     val chainJob = async { runKMathChained() }
+    val blockingJob = async { runKMathBlocking() }
     println("KMath Chained: ${chainJob.await()}")
+    println("KMath Blocking: ${blockingJob.await()}")
     println("Apache Direct: ${directJob.await()}")
 }

kmath-core/src/commonMain/kotlin/space/kscience/kmath/linear/MatrixBuilder.kt

@@ -67,13 +67,14 @@ public fun <T : Any, A : Ring<T>> MatrixBuilder<T, A>.symmetric(
 ): Matrix<T> {
     require(columns == rows) { "In order to build symmetric matrix, number of rows $rows should be equal to number of columns $columns" }
     return with(BufferAccessor2D<T?>(rows, rows, MutableBufferFactory(type))) {
-        val cache = factory(rows * rows) { null }
+        val cache = HashMap<IntArray, T>()
         linearSpace.buildMatrix(rows, rows) { i, j ->
-            val cached = cache[i, j]
+            val index = intArrayOf(i, j)
+            val cached = cache[index]
             if (cached == null) {
                 val value = if (i <= j) builder(i, j) else builder(j, i)
-                cache[i, j] = value
+                cache[index] = value
-                cache[j, i] = value
+                cache[index] = value
                 value
             } else {
                 cached