[SPARK-15569] Reduce frequency of updateBytesWritten function in Disk…

## What changes were proposed in this pull request?

Profiling a Spark job spilling large amount of intermediate data we found that significant portion of time is being spent in DiskObjectWriter.updateBytesWritten function. Looking at the code, we see that the function is being called too frequently to update the number of bytes written to disk. We should reduce the frequency to avoid this.

## How was this patch tested?

Tested by running the job on cluster and saw 20% CPU gain  by this change.

Author: Sital Kedia <[email protected]>

Closes apache#13332 from sitalkedia/DiskObjectWriter.

Author: Sital Kedia

core/src/main/scala/org/apache/spark/storage/DiskBlockObjectWriter.scala

@@ -203,8 +203,7 @@ private[spark] class DiskBlockObjectWriter(
     numRecordsWritten += 1
     writeMetrics.incRecordsWritten(1)
 
-    // TODO: call updateBytesWritten() less frequently.
-    if (numRecordsWritten % 32 == 0) {
+    if (numRecordsWritten % 16384 == 0) {
       updateBytesWritten()
     }
   }

core/src/test/scala/org/apache/spark/storage/DiskBlockObjectWriterSuite.scala

@@ -53,13 +53,13 @@ class DiskBlockObjectWriterSuite extends SparkFunSuite with BeforeAndAfterEach {
     assert(writeMetrics.recordsWritten === 1)
     // Metrics don't update on every write
     assert(writeMetrics.bytesWritten == 0)
-    // After 32 writes, metrics should update
-    for (i <- 0 until 32) {
+    // After 16384 writes, metrics should update
+    for (i <- 0 until 16384) {
       writer.flush()
       writer.write(Long.box(i), Long.box(i))
     }
     assert(writeMetrics.bytesWritten > 0)
-    assert(writeMetrics.recordsWritten === 33)
+    assert(writeMetrics.recordsWritten === 16385)
     writer.commitAndClose()
     assert(file.length() == writeMetrics.bytesWritten)
   }
@@ -75,13 +75,13 @@ class DiskBlockObjectWriterSuite extends SparkFunSuite with BeforeAndAfterEach {
     assert(writeMetrics.recordsWritten === 1)
     // Metrics don't update on every write
     assert(writeMetrics.bytesWritten == 0)
-    // After 32 writes, metrics should update
-    for (i <- 0 until 32) {
+    // After 16384 writes, metrics should update
+    for (i <- 0 until 16384) {
       writer.flush()
       writer.write(Long.box(i), Long.box(i))
     }
     assert(writeMetrics.bytesWritten > 0)
-    assert(writeMetrics.recordsWritten === 33)
+    assert(writeMetrics.recordsWritten === 16385)
     writer.revertPartialWritesAndClose()
     assert(writeMetrics.bytesWritten == 0)
     assert(writeMetrics.recordsWritten == 0)