From 52834d761b059264214dfc6a1f9c70b8bc7ec089 Mon Sep 17 00:00:00 2001
From: Aaron Davidson <aaron@databricks.com>
Date: Sun, 9 Mar 2014 11:08:39 -0700
Subject: SPARK-929: Fully deprecate usage of SPARK_MEM

(Continued from old repo, prior discussion at https://github.com/apache/incubator-spark/pull/615)

This patch cements our deprecation of the SPARK_MEM environment variable by replacing it with three more specialized variables:
SPARK_DAEMON_MEMORY, SPARK_EXECUTOR_MEMORY, and SPARK_DRIVER_MEMORY

The creation of the latter two variables means that we can safely set driver/job memory without accidentally setting the executor memory. Neither is public.

SPARK_EXECUTOR_MEMORY is only used by the Mesos scheduler (and set within SparkContext). The proper way of configuring executor memory is through the "spark.executor.memory" property.

SPARK_DRIVER_MEMORY is the new way of specifying the amount of memory run by jobs launched by spark-class, without possibly affecting executor memory.

Other memory considerations:
- The repl's memory can be set through the "--drivermem" command-line option, which really just sets SPARK_DRIVER_MEMORY.
- run-example doesn't use spark-class, so the only way to modify examples' memory is actually an unusual use of SPARK_JAVA_OPTS (which is normally overriden in all cases by spark-class).

This patch also fixes a lurking bug where spark-shell misused spark-class (the first argument is supposed to be the main class name, not java options), as well as a bug in the Windows spark-class2.cmd. I have not yet tested this patch on either Windows or Mesos, however.

Author: Aaron Davidson <aaron@databricks.com>

Closes #99 from aarondav/sparkmem and squashes the following commits:

9df4c68 [Aaron Davidson] SPARK-929: Fully deprecate usage of SPARK_MEM
---
 .../main/scala/org/apache/spark/SparkContext.scala   | 20 +++++++++++---------
 .../src/main/scala/org/apache/spark/util/Utils.scala |  2 --
 2 files changed, 11 insertions(+), 11 deletions(-)

(limited to 'core')

diff --git a/core/src/main/scala/org/apache/spark/SparkContext.scala b/core/src/main/scala/org/apache/spark/SparkContext.scala
index ce25573834..cdc0e5a342 100644
--- a/core/src/main/scala/org/apache/spark/SparkContext.scala
+++ b/core/src/main/scala/org/apache/spark/SparkContext.scala
@@ -162,19 +162,20 @@ class SparkContext(
     jars.foreach(addJar)
   }
 
+  def warnSparkMem(value: String): String = {
+    logWarning("Using SPARK_MEM to set amount of memory to use per executor process is " +
+      "deprecated, please use spark.executor.memory instead.")
+    value
+  }
+
   private[spark] val executorMemory = conf.getOption("spark.executor.memory")
-    .orElse(Option(System.getenv("SPARK_MEM")))
+    .orElse(Option(System.getenv("SPARK_EXECUTOR_MEMORY")))
+    .orElse(Option(System.getenv("SPARK_MEM")).map(warnSparkMem))
     .map(Utils.memoryStringToMb)
     .getOrElse(512)
 
-  if (!conf.contains("spark.executor.memory") && sys.env.contains("SPARK_MEM")) {
-    logWarning("Using SPARK_MEM to set amount of memory to use per executor process is " +
-      "deprecated, instead use spark.executor.memory")
-  }
-
   // Environment variables to pass to our executors
   private[spark] val executorEnvs = HashMap[String, String]()
-  // Note: SPARK_MEM is included for Mesos, but overwritten for standalone mode in ExecutorRunner
   for (key <- Seq("SPARK_CLASSPATH", "SPARK_LIBRARY_PATH", "SPARK_JAVA_OPTS");
       value <- Option(System.getenv(key))) {
     executorEnvs(key) = value
@@ -185,8 +186,9 @@ class SparkContext(
     value <- Option(System.getenv(envKey)).orElse(Option(System.getProperty(propKey)))} {
     executorEnvs(envKey) = value
   }
-  // Since memory can be set with a system property too, use that
-  executorEnvs("SPARK_MEM") = executorMemory + "m"
+  // The Mesos scheduler backend relies on this environment variable to set executor memory.
+  // TODO: Set this only in the Mesos scheduler.
+  executorEnvs("SPARK_EXECUTOR_MEMORY") = executorMemory + "m"
   executorEnvs ++= conf.getExecutorEnv
 
   // Set SPARK_USER for user who is running SparkContext.
diff --git a/core/src/main/scala/org/apache/spark/util/Utils.scala b/core/src/main/scala/org/apache/spark/util/Utils.scala
index 0eb2f78b73..53458b6660 100644
--- a/core/src/main/scala/org/apache/spark/util/Utils.scala
+++ b/core/src/main/scala/org/apache/spark/util/Utils.scala
@@ -532,8 +532,6 @@ private[spark] object Utils extends Logging {
 
   /**
    * Convert a Java memory parameter passed to -Xmx (such as 300m or 1g) to a number of megabytes.
-   * This is used to figure out how much memory to claim from Mesos based on the SPARK_MEM
-   * environment variable.
    */
   def memoryStringToMb(str: String): Int = {
     val lower = str.toLowerCase
-- 
cgit v1.2.3