feat(capelin): Use logical types for Parquet output columns

This change updates the output schema for the experiment data to use
logical types where possible. This adds additional context for the
writer and the reader on how to process the column (efficiently).

4 files changed, 63 insertions, 16 deletions

diff --git a/opendc-experiments/opendc-experiments-capelin/src/main/kotlin/org/opendc/experiments/capelin/export/parquet/AvroUtils.kt b/opendc-experiments/opendc-experiments-capelin/src/main/kotlin/org/opendc/experiments/capelin/export/parquet/AvroUtils.kt
new file mode 100644
index 00000000..a4676f31
--- /dev/null
+++ b/opendc-experiments/opendc-experiments-capelin/src/main/kotlin/org/opendc/experiments/capelin/export/parquet/AvroUtils.kt
@@ -0,0 +1,44 @@
+/*
+ * 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.
+ */
+
+@file:JvmName("AvroUtils")
+package org.opendc.experiments.capelin.export.parquet
+
+import org.apache.avro.LogicalTypes
+import org.apache.avro.Schema
+
+/**
+ * Schema for UUID type.
+ */
+internal val UUID_SCHEMA = LogicalTypes.uuid().addToSchema(Schema.create(Schema.Type.STRING))
+
+/**
+ * Schema for timestamp type.
+ */
+internal val TIMESTAMP_SCHEMA = LogicalTypes.timestampMillis().addToSchema(Schema.create(Schema.Type.LONG))
+
+/**
+ * Helper function to make a [Schema] field optional.
+ */
+internal fun Schema.optional(): Schema {
+    return Schema.createUnion(Schema.create(Schema.Type.NULL), this)
+}
diff --git a/opendc-experiments/opendc-experiments-capelin/src/main/kotlin/org/opendc/experiments/capelin/export/parquet/ParquetHostDataWriter.kt b/opendc-experiments/opendc-experiments-capelin/src/main/kotlin/org/opendc/experiments/capelin/export/parquet/ParquetHostDataWriter.kt
index 36207045..58388cb1 100644
--- a/opendc-experiments/opendc-experiments-capelin/src/main/kotlin/org/opendc/experiments/capelin/export/parquet/ParquetHostDataWriter.kt
+++ b/opendc-experiments/opendc-experiments-capelin/src/main/kotlin/org/opendc/experiments/capelin/export/parquet/ParquetHostDataWriter.kt
@@ -47,8 +47,6 @@ public class ParquetHostDataWriter(path: File, bufferSize: Int) :
         builder["timestamp"] = data.timestamp.toEpochMilli()
 
         builder["host_id"] = data.host.id
-        builder["num_cpus"] = data.host.cpuCount
-        builder["mem_capacity"] = data.host.memCapacity
 
         builder["uptime"] = data.uptime
         builder["downtime"] = data.downtime
@@ -57,12 +55,15 @@ public class ParquetHostDataWriter(path: File, bufferSize: Int) :
             builder["boot_time"] = bootTime.toEpochMilli()
         }
 
+        builder["cpu_count"] = data.host.cpuCount
         builder["cpu_limit"] = data.cpuLimit
         builder["cpu_time_active"] = data.cpuActiveTime
         builder["cpu_time_idle"] = data.cpuIdleTime
         builder["cpu_time_steal"] = data.cpuStealTime
         builder["cpu_time_lost"] = data.cpuLostTime
 
+        builder["mem_limit"] = data.host.memCapacity
+
         builder["power_total"] = data.powerTotal
 
         builder["guests_terminated"] = data.guestsTerminated
@@ -78,18 +79,18 @@ public class ParquetHostDataWriter(path: File, bufferSize: Int) :
             .record("host")
             .namespace("org.opendc.telemetry.compute")
             .fields()
-            .requiredLong("timestamp")
-            .requiredString("host_id")
-            .requiredInt("num_cpus")
-            .requiredLong("mem_capacity")
+            .name("timestamp").type(TIMESTAMP_SCHEMA).noDefault()
+            .name("host_id").type(UUID_SCHEMA).noDefault()
             .requiredLong("uptime")
             .requiredLong("downtime")
-            .optionalLong("boot_time")
+            .name("boot_time").type(TIMESTAMP_SCHEMA.optional()).noDefault()
+            .requiredInt("cpu_count")
             .requiredDouble("cpu_limit")
             .requiredLong("cpu_time_active")
             .requiredLong("cpu_time_idle")
             .requiredLong("cpu_time_steal")
             .requiredLong("cpu_time_lost")
+            .requiredLong("mem_limit")
             .requiredDouble("power_total")
             .requiredInt("guests_terminated")
             .requiredInt("guests_running")
diff --git a/opendc-experiments/opendc-experiments-capelin/src/main/kotlin/org/opendc/experiments/capelin/export/parquet/ParquetServerDataWriter.kt b/opendc-experiments/opendc-experiments-capelin/src/main/kotlin/org/opendc/experiments/capelin/export/parquet/ParquetServerDataWriter.kt
index c5a5e7c0..43b5f469 100644
--- a/opendc-experiments/opendc-experiments-capelin/src/main/kotlin/org/opendc/experiments/capelin/export/parquet/ParquetServerDataWriter.kt
+++ b/opendc-experiments/opendc-experiments-capelin/src/main/kotlin/org/opendc/experiments/capelin/export/parquet/ParquetServerDataWriter.kt
@@ -30,6 +30,7 @@ import org.apache.parquet.avro.AvroParquetWriter
 import org.apache.parquet.hadoop.ParquetWriter
 import org.opendc.telemetry.compute.table.ServerData
 import java.io.File
+import java.util.*
 
 /**
  * A Parquet event writer for [ServerData]s.
@@ -49,8 +50,6 @@ public class ParquetServerDataWriter(path: File, bufferSize: Int) :
 
         builder["server_id"] = data.server.id
         builder["host_id"] = data.host?.id
-        builder["num_vcpus"] = data.server.cpuCount
-        builder["mem_capacity"] = data.server.memCapacity
 
         builder["uptime"] = data.uptime
         builder["downtime"] = data.downtime
@@ -60,11 +59,14 @@ public class ParquetServerDataWriter(path: File, bufferSize: Int) :
         }
         builder["scheduling_latency"] = data.schedulingLatency
 
+        builder["cpu_count"] = data.server.cpuCount
         builder["cpu_limit"] = data.cpuLimit
         builder["cpu_time_active"] = data.cpuActiveTime
         builder["cpu_time_idle"] = data.cpuIdleTime
         builder["cpu_time_steal"] = data.cpuStealTime
         builder["cpu_time_lost"] = data.cpuLostTime
+
+        builder["mem_limit"] = data.server.memCapacity
     }
 
     override fun toString(): String = "server-writer"
@@ -74,20 +76,20 @@ public class ParquetServerDataWriter(path: File, bufferSize: Int) :
             .record("server")
             .namespace("org.opendc.telemetry.compute")
             .fields()
-            .requiredLong("timestamp")
-            .requiredString("server_id")
-            .optionalString("host_id")
-            .requiredInt("num_vcpus")
-            .requiredLong("mem_capacity")
+            .name("timestamp").type(TIMESTAMP_SCHEMA).noDefault()
+            .name("server_id").type(UUID_SCHEMA).noDefault()
+            .name("host_id").type(UUID_SCHEMA.optional()).noDefault()
             .requiredLong("uptime")
             .requiredLong("downtime")
-            .optionalLong("boot_time")
+            .name("boot_time").type(TIMESTAMP_SCHEMA.optional()).noDefault()
             .requiredLong("scheduling_latency")
+            .requiredInt("cpu_count")
             .requiredDouble("cpu_limit")
             .requiredLong("cpu_time_active")
             .requiredLong("cpu_time_idle")
             .requiredLong("cpu_time_steal")
             .requiredLong("cpu_time_lost")
+            .requiredLong("mem_limit")
             .endRecord()
     }
 }
diff --git a/opendc-experiments/opendc-experiments-capelin/src/main/kotlin/org/opendc/experiments/capelin/export/parquet/ParquetServiceDataWriter.kt b/opendc-experiments/opendc-experiments-capelin/src/main/kotlin/org/opendc/experiments/capelin/export/parquet/ParquetServiceDataWriter.kt
index d9ca55cb..2928f445 100644
--- a/opendc-experiments/opendc-experiments-capelin/src/main/kotlin/org/opendc/experiments/capelin/export/parquet/ParquetServiceDataWriter.kt
+++ b/opendc-experiments/opendc-experiments-capelin/src/main/kotlin/org/opendc/experiments/capelin/export/parquet/ParquetServiceDataWriter.kt
@@ -52,7 +52,7 @@ public class ParquetServiceDataWriter(path: File, bufferSize: Int) :
             .record("service")
             .namespace("org.opendc.telemetry.compute")
             .fields()
-            .requiredLong("timestamp")
+            .name("timestamp").type(TIMESTAMP_SCHEMA).noDefault()
             .requiredInt("hosts_up")
             .requiredInt("hosts_down")
             .requiredInt("servers_pending")