[GitHub] [arrow] zhztheplayer commented on a change in pull request #7030: ARROW-7808: [Java][Dataset] Implement Datasets Java API by JNI to C++

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

zhztheplayer commented on a change in pull request #7030:
URL: https://github.com/apache/arrow/pull/7030#discussion_r449956827



##########
File path: java/dataset/src/main/java/org/apache/arrow/memory/NativeUnderlingMemory.java
##########
@@ -0,0 +1,65 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *    http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.arrow.memory;
+
+import org.apache.arrow.dataset.jni.JniWrapper;
+
+/**
+ * AllocationManager implementation for Native allocated memory.
+ */
+public class NativeUnderlingMemory extends AllocationManager {

Review comment:
       > In reality, this is not at all the case. There are two ways that you run out of memory in Java: you run out of available allocation space OR you hit the limit at the JVM level. The limit at the JVM level can be influenced by things like fragmentation so in nearly all cases you hit that limit before the allocator managed limit. In your patch, you only are respecting the allocator set limit, not the jvm limit. This means a user could configure java to allow X of memory and the process actually will use 2X of memory, leading to issues in many systems.
   
   Making sense overall. So we may have to use a C++-to-Java/Netty allocation path in the first version right? I'll try it out.




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to 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