Github user sameeragarwal commented on a diff in the pull request:
https://github.com/apache/spark/pull/14174#discussion_r71924045
--- Diff:
sql/catalyst/src/main/java/org/apache/spark/sql/catalyst/expressions/RowBasedKeyValueBatch.java
---
@@ -0,0 +1,182 @@
+/*
+ * 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.spark.sql.catalyst.expressions;
+
+import java.io.IOException;
+
+import org.apache.spark.memory.MemoryConsumer;
+import org.apache.spark.memory.TaskMemoryManager;
+import org.apache.spark.sql.types.*;
+import org.apache.spark.unsafe.memory.MemoryBlock;
+
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+
+/**
+ * RowBasedKeyValueBatch stores key value pairs in contiguous memory
region.
+ *
+ * Each key or value is stored as a single UnsafeRow. Each record contains
one key and one value
+ * and some auxiliary data, which differs based on implementation:
+ * i.e., `FixedLengthRowBasedKeyValueBatch` and
`VariableLengthRowBasedKeyValueBatch`.
+ *
+ * We use `FixedLengthRowBasedKeyValueBatch` if all fiends in the key and
the value are fixed-length
+ * data types. Otherwise we use `VariableLengthRowBasedKeyValueBatch`.
+ *
+ * RowBasedKeyValueBatch is backed by a single page / MemoryBlock
(defaults to 64MB). If the page
+ * is full, the aggregate logic should fallback to a second level, larger
hash map. We intentionally
+ * use the single-page design because it simplifies memory address
encoding & decoding for each
+ * key-value pair. Because the maximum capacity for RowBasedKeyValueBatch
is only 2^16, it is
+ * unlikely we need a second page anyway. Filling the page requires an
average size for key value
+ * pairs to be larger than 1024 bytes.
+ *
+ */
+public abstract class RowBasedKeyValueBatch extends MemoryConsumer {
+ protected final Logger logger =
LoggerFactory.getLogger(RowBasedKeyValueBatch.class);
+
+ protected static final int DEFAULT_CAPACITY = 1 << 16;
--- End diff --
This and `DEFAULT_PAGE_SIZE` can probably be private?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
