HADOOP-6502. Improve the performance of Configuration.getClassByName when the class is not found by caching negative results. Contributed by Sharad Agarwal and Todd Lipcon.

git-svn-id: https://svn.apache.org/repos/asf/hadoop/common/branches/branch-0.23@1244619 13f79535-47bb-0310-9956-ffa450edef68

hadoop-common-project/hadoop-common/CHANGES.txt

@@ -17,6 +17,10 @@ Release 0.23.2 - UNRELEASED
     HADOOP-8071. Avoid an extra packet in client code when nagling is
     disabled. (todd)
 
+    HADOOP-6502. Improve the performance of Configuration.getClassByName when
+    the class is not found by caching negative results.
+    (sharad, todd via todd)
+
   BUG FIXES
 
     HADOOP-8042  When copying a file out of HDFS, modifying it, and uploading

hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/conf/Configuration.java

@@ -1146,6 +1146,22 @@ public class Configuration implements Iterable<Map.Entry<String,String>>,
    * @throws ClassNotFoundException if the class is not found.
    */
   public Class<?> getClassByName(String name) throws ClassNotFoundException {
+    Class<?> ret = getClassByNameOrNull(name);
+    if (ret == null) {
+      throw new ClassNotFoundException("Class " + name + " not found");
+    }
+    return ret;
+  }
+  
+  /**
+   * Load a class by name, returning null rather than throwing an exception
+   * if it couldn't be loaded. This is to avoid the overhead of creating
+   * an exception.
+   * 
+   * @param name the class name
+   * @return the class object, or null if it could not be found.
+   */
+  public Class<?> getClassByNameOrNull(String name) {
     Map<String, Class<?>> map;
     
     synchronized (CACHE_CLASSES) {
@@ -1157,12 +1173,20 @@ public class Configuration implements Iterable<Map.Entry<String,String>>,
       }
     }
 
-    Class<?> clazz = map.get(name);
+    Class<?> clazz = null;
-    if (clazz == null) {
+    if (!map.containsKey(name)) {
-      clazz = Class.forName(name, true, classLoader);
+      try {
-      if (clazz != null) {
+        clazz = Class.forName(name, true, classLoader);
-        // two putters can race here, but they'll put the same class
+      } catch (ClassNotFoundException e) {
-        map.put(name, clazz);
+        map.put(name, null); //cache negative that class is not found
+        return null;
+      }
+      // two putters can race here, but they'll put the same class
+      map.put(name, clazz);
+    } else { // check already performed on this class name
+      clazz = map.get(name);
+      if (clazz == null) { // found the negative
+        return null;
       }
     }
 

hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/util/ReflectionUtils.java

@@ -86,17 +86,22 @@ public class ReflectionUtils {
     //invoke configure on theObject
     try {
       Class<?> jobConfClass = 
-        conf.getClassByName("org.apache.hadoop.mapred.JobConf");
+        conf.getClassByNameOrNull("org.apache.hadoop.mapred.JobConf");
+      if (jobConfClass == null) {
+        return;
+      }
+      
       Class<?> jobConfigurableClass = 
-        conf.getClassByName("org.apache.hadoop.mapred.JobConfigurable");
+        conf.getClassByNameOrNull("org.apache.hadoop.mapred.JobConfigurable");
-       if (jobConfClass.isAssignableFrom(conf.getClass()) &&
+      if (jobConfigurableClass == null) {
+        return;
+      }
+      if (jobConfClass.isAssignableFrom(conf.getClass()) &&
             jobConfigurableClass.isAssignableFrom(theObject.getClass())) {
         Method configureMethod = 
           jobConfigurableClass.getMethod("configure", jobConfClass);
         configureMethod.invoke(theObject, conf);
       }
-    } catch (ClassNotFoundException e) {
-      //JobConf/JobConfigurable not in classpath. no need to configure
     } catch (Exception e) {
       throw new RuntimeException("Error in configuring object", e);
     }