[GitHub] squito commented on a change in pull request #23688: [SPARK-25035][Core] Avoiding memory mapping at disk-stored blocks replication

[GitBox <gi...@apache.org>]

squito commented on a change in pull request #23688: [SPARK-25035][Core] Avoiding memory mapping at disk-stored blocks replication
URL: https://github.com/apache/spark/pull/23688#discussion_r254073272
 
 

 ##########
 File path: core/src/test/scala/org/apache/spark/storage/BlockManagerSuite.scala
 ##########
 @@ -906,6 +910,55 @@ class BlockManagerSuite extends SparkFunSuite with Matchers with BeforeAndAfterE
     }
   }
 
+  def testPutBlockDataAsStream(
+      name: String,
+      blockManager: BlockManager,
+      storageLevel: StorageLevel): Unit = {
+    val message = "message"
+    val ser = serializer.newInstance().serialize(message).array()
+    val blockId = new ShuffleDataBlockId(0, 0, 0)
+    val streamCallbackWithId =
+      blockManager.putBlockDataAsStream(blockId, storageLevel, ClassTag(message.getClass))
+    streamCallbackWithId.onData("0", ByteBuffer.wrap(ser))
+    streamCallbackWithId.onComplete("0")
+    val blockStatusOption = blockManager.getStatus(blockId)
+    assert(!blockStatusOption.isEmpty)
+    val blockStatus = blockStatusOption.get
+    assert((blockStatus.diskSize > 0) === !storageLevel.useMemory)
+    assert((blockStatus.memSize > 0) === storageLevel.useMemory)
+    assert(blockManager.getBlockData(blockId).nioByteBuffer().array() === ser)
+  }
+
+  Seq(
+    "caching" -> StorageLevel.MEMORY_ONLY,
+    "caching on disk" -> StorageLevel.DISK_ONLY,
+    "caching in memory, replicated" -> StorageLevel.MEMORY_ONLY_2,
+    "caching in memory, serialized, replicated" -> StorageLevel.MEMORY_ONLY_SER_2,
+    "caching on disk, replicated" -> StorageLevel.DISK_ONLY_2,
+    "caching in memory and disk, replicated" -> StorageLevel.MEMORY_AND_DISK_2,
+    "caching in memory and disk, serialized, replicated" -> StorageLevel.MEMORY_AND_DISK_SER_2
 
 Review comment:
   you've only setup one BlockManager here, so is there any point in running the tests w/ replication? Replication will be covered by the tests in DistributedSuite.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org