summaryrefslogtreecommitdiff
path: root/opendc-format/src/main
diff options
context:
space:
mode:
authorFabian Mastenbroek <mail.fabianm@gmail.com>2021-08-24 13:15:26 +0200
committerGitHub <noreply@github.com>2021-08-24 13:15:26 +0200
commitac48fa12f36180de31154a7c828b4dc281dac94b (patch)
tree3a1117b62e627094ea2859a8b1cb910ef7046851 /opendc-format/src/main
parent51515bb255b3b32ca3020419a0c84130a4d8d370 (diff)
parent5266ecd476a18f601cb4eb6166f4c8338c440210 (diff)
merge: Add tests for interference and failures in Capelin
This pull request updates the Capelin experiments to test for interference and failure scenarios. This allows us to track regressions in these subsystems more easily. * Clean up Bitbrains trace reader to enable re-use * Keep trace order after sampling * Update Bitbrains trace tests * Add support for reporting interfered work * Add support for SimHost failure * Add tests for interference and failures
Diffstat (limited to 'opendc-format/src/main')
-rw-r--r--opendc-format/src/main/kotlin/org/opendc/format/trace/bitbrains/BitbrainsRawTraceReader.kt100
-rw-r--r--opendc-format/src/main/kotlin/org/opendc/format/trace/bitbrains/BitbrainsTraceReader.kt97
2 files changed, 139 insertions, 58 deletions
diff --git a/opendc-format/src/main/kotlin/org/opendc/format/trace/bitbrains/BitbrainsRawTraceReader.kt b/opendc-format/src/main/kotlin/org/opendc/format/trace/bitbrains/BitbrainsRawTraceReader.kt
new file mode 100644
index 00000000..ff6cdd02
--- /dev/null
+++ b/opendc-format/src/main/kotlin/org/opendc/format/trace/bitbrains/BitbrainsRawTraceReader.kt
@@ -0,0 +1,100 @@
+/*
+ * Copyright (c) 2021 AtLarge Research
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in all
+ * copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+package org.opendc.format.trace.bitbrains
+
+import com.fasterxml.jackson.annotation.JsonProperty
+import com.fasterxml.jackson.databind.MappingIterator
+import com.fasterxml.jackson.dataformat.csv.CsvMapper
+import com.fasterxml.jackson.dataformat.csv.CsvSchema
+import java.io.InputStream
+
+/**
+ * A trace reader that enables the user to read Bitbrains specific trace data.
+ */
+public class BitbrainsRawTraceReader(input: InputStream) : Iterator<BitbrainsRawTraceReader.Entry>, AutoCloseable {
+ /**
+ * The [CsvSchema] that is used to parse the trace.
+ */
+ private val schema = CsvSchema.builder()
+ .addColumn("Timestamp [ms]", CsvSchema.ColumnType.NUMBER)
+ .addColumn("CPU cores", CsvSchema.ColumnType.NUMBER)
+ .addColumn("CPU capacity provisioned [MHZ]", CsvSchema.ColumnType.NUMBER)
+ .addColumn("CPU usage [MHZ]", CsvSchema.ColumnType.NUMBER)
+ .addColumn("CPU usage [%]", CsvSchema.ColumnType.NUMBER)
+ .addColumn("Memory capacity provisioned [KB]", CsvSchema.ColumnType.NUMBER)
+ .addColumn("Memory usage [KB]", CsvSchema.ColumnType.NUMBER)
+ .addColumn("Disk read throughput [KB/s]", CsvSchema.ColumnType.NUMBER)
+ .addColumn("Disk write throughput [KB/s]", CsvSchema.ColumnType.NUMBER)
+ .addColumn("Network received throughput [KB/s]", CsvSchema.ColumnType.NUMBER)
+ .addColumn("Network transmitted throughput [KB/s]", CsvSchema.ColumnType.NUMBER)
+ .setAllowComments(true)
+ .setUseHeader(true)
+ .setColumnSeparator(';')
+ .build()
+
+ /**
+ * The mapping iterator to use.
+ */
+ private val iterator: MappingIterator<Entry> = CsvMapper().readerFor(Entry::class.java).with(schema)
+ .readValues(input)
+
+ override fun hasNext(): Boolean {
+ return iterator.hasNext()
+ }
+
+ override fun next(): Entry {
+ return iterator.next()
+ }
+
+ override fun close() {
+ iterator.close()
+ }
+
+ /**
+ * A single entry in the trace.
+ */
+ public data class Entry(
+ @JsonProperty("Timestamp [ms]")
+ val timestamp: Long,
+ @JsonProperty("CPU cores")
+ val cpuCores: Int,
+ @JsonProperty("CPU capacity provisioned [MHZ]")
+ val cpuCapacity: Double,
+ @JsonProperty("CPU usage [MHZ]")
+ val cpuUsage: Double,
+ @JsonProperty("CPU usage [%]")
+ val cpuUsagePct: Double,
+ @JsonProperty("Memory capacity provisioned [KB]")
+ val memCapacity: Double,
+ @JsonProperty("Memory usage [KB]")
+ val memUsage: Double,
+ @JsonProperty("Disk read throughput [KB/s]")
+ val diskRead: Double,
+ @JsonProperty("Disk write throughput [KB/s]")
+ val diskWrite: Double,
+ @JsonProperty("Network received throughput [KB/s]")
+ val netReceived: Double,
+ @JsonProperty("Network transmitted throughput [KB/s]")
+ val netTransmitted: Double
+ )
+}
diff --git a/opendc-format/src/main/kotlin/org/opendc/format/trace/bitbrains/BitbrainsTraceReader.kt b/opendc-format/src/main/kotlin/org/opendc/format/trace/bitbrains/BitbrainsTraceReader.kt
index cd8021fe..9e4876df 100644
--- a/opendc-format/src/main/kotlin/org/opendc/format/trace/bitbrains/BitbrainsTraceReader.kt
+++ b/opendc-format/src/main/kotlin/org/opendc/format/trace/bitbrains/BitbrainsTraceReader.kt
@@ -26,10 +26,10 @@ import org.opendc.format.trace.TraceEntry
import org.opendc.format.trace.TraceReader
import org.opendc.simulator.compute.workload.SimTraceWorkload
import org.opendc.simulator.compute.workload.SimWorkload
-import java.io.BufferedReader
import java.io.File
-import java.io.FileReader
+import java.io.FileInputStream
import java.util.*
+import kotlin.math.max
import kotlin.math.min
/**
@@ -48,74 +48,55 @@ public class BitbrainsTraceReader(traceDirectory: File) : TraceReader<SimWorkloa
*/
init {
val entries = mutableMapOf<Long, TraceEntry<SimWorkload>>()
-
- var timestampCol = 0
- var coreCol = 0
- var cpuUsageCol = 0
- var provisionedMemoryCol = 0
val traceInterval = 5 * 60 * 1000L
traceDirectory.walk()
.filterNot { it.isDirectory }
+ .filter { it.extension == "csv" }
.forEach { vmFile ->
- println(vmFile)
val flopsHistory = mutableListOf<SimTraceWorkload.Fragment>()
var vmId = -1L
- var cores = -1
- var requiredMemory = -1L
- var startTime = -1L
-
- BufferedReader(FileReader(vmFile)).use { reader ->
- reader.lineSequence()
- .filter { line ->
- // Ignore comments in the trace
- !line.startsWith("#") && line.isNotBlank()
+ var maxCores = Int.MIN_VALUE
+ var requiredMemory = Long.MIN_VALUE
+ var startTime = Long.MAX_VALUE
+ var lastTimestamp = Long.MIN_VALUE
+
+ BitbrainsRawTraceReader(FileInputStream(vmFile)).use { reader ->
+ reader.forEach { entry ->
+ val timestamp = entry.timestamp * 1000L
+ val cpuUsage = entry.cpuUsage
+ vmId = vmFile.nameWithoutExtension.trim().toLong()
+ val cores = entry.cpuCores
+ maxCores = max(maxCores, cores)
+ requiredMemory = max(requiredMemory, (entry.memCapacity / 1000).toLong())
+
+ if (lastTimestamp < 0) {
+ lastTimestamp = timestamp - 5 * 60 * 1000L
+ startTime = min(startTime, lastTimestamp)
}
- .forEachIndexed { idx, line ->
- val values = line.split(";\t")
-
- // Parse GWF header
- if (idx == 0) {
- val header = values.mapIndexed { col, name -> Pair(name.trim(), col) }.toMap()
- timestampCol = header["Timestamp [ms]"]!!
- coreCol = header["CPU cores"]!!
- cpuUsageCol = header["CPU usage [MHZ]"]!!
- provisionedMemoryCol = header["Memory capacity provisioned [KB]"]!!
- return@forEachIndexed
- }
- vmId = vmFile.nameWithoutExtension.trim().toLong()
- val timestamp = values[timestampCol].trim().toLong() - 5 * 60
- startTime = min(startTime, timestamp)
- cores = values[coreCol].trim().toInt()
- val cpuUsage = values[cpuUsageCol].trim().toDouble() // MHz
- requiredMemory = (values[provisionedMemoryCol].trim().toDouble() / 1000).toLong()
-
- if (flopsHistory.isEmpty()) {
- flopsHistory.add(SimTraceWorkload.Fragment(timestamp, traceInterval, cpuUsage, cores))
+ if (flopsHistory.isEmpty()) {
+ flopsHistory.add(SimTraceWorkload.Fragment(lastTimestamp, traceInterval, cpuUsage, cores))
+ } else {
+ val last = flopsHistory.last()
+ val duration = timestamp - lastTimestamp
+ // Perform run-length encoding
+ if (duration == 0L || last.usage == cpuUsage) {
+ flopsHistory[flopsHistory.size - 1] = last.copy(duration = last.duration + duration)
} else {
- if (flopsHistory.last().usage != cpuUsage) {
- flopsHistory.add(
- SimTraceWorkload.Fragment(
- timestamp,
- traceInterval,
- cpuUsage,
- cores
- )
- )
- } else {
- val oldFragment = flopsHistory.removeAt(flopsHistory.size - 1)
- flopsHistory.add(
- SimTraceWorkload.Fragment(
- oldFragment.timestamp,
- oldFragment.duration + traceInterval,
- cpuUsage,
- cores
- )
+ flopsHistory.add(
+ SimTraceWorkload.Fragment(
+ lastTimestamp,
+ duration,
+ cpuUsage,
+ cores
)
- }
+ )
}
}
+
+ lastTimestamp = timestamp
+ }
}
val uuid = UUID(0L, vmId)
@@ -127,7 +108,7 @@ public class BitbrainsTraceReader(traceDirectory: File) : TraceReader<SimWorkloa
startTime,
workload,
mapOf(
- "cores" to cores,
+ "cores" to maxCores,
"required-memory" to requiredMemory,
"workload" to workload
)