Github user paul-rogers commented on a diff in the pull request:
https://github.com/apache/drill/pull/717#discussion_r99067830
--- Diff:
exec/java-exec/src/main/java/org/apache/drill/exec/physical/impl/xsort/managed/ExternalSortBatch.java
---
@@ -0,0 +1,1321 @@
+/*
+ * 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.drill.exec.physical.impl.xsort.managed;
+
+import java.io.IOException;
+import java.util.Collection;
+import java.util.LinkedList;
+import java.util.List;
+
+import org.apache.drill.common.AutoCloseables;
+import org.apache.drill.common.config.DrillConfig;
+import org.apache.drill.common.exceptions.UserException;
+import org.apache.drill.exec.ExecConstants;
+import org.apache.drill.exec.exception.OutOfMemoryException;
+import org.apache.drill.exec.exception.SchemaChangeException;
+import org.apache.drill.exec.memory.BufferAllocator;
+import org.apache.drill.exec.ops.FragmentContext;
+import org.apache.drill.exec.ops.MetricDef;
+import org.apache.drill.exec.physical.config.ExternalSort;
+import org.apache.drill.exec.physical.impl.sort.RecordBatchData;
+import org.apache.drill.exec.physical.impl.xsort.MSortTemplate;
+import org.apache.drill.exec.physical.impl.xsort.SingleBatchSorter;
+import
org.apache.drill.exec.physical.impl.xsort.managed.BatchGroup.InputBatch;
+import org.apache.drill.exec.record.AbstractRecordBatch;
+import org.apache.drill.exec.record.BatchSchema;
+import org.apache.drill.exec.record.BatchSchema.SelectionVectorMode;
+import org.apache.drill.exec.record.RecordBatch;
+import org.apache.drill.exec.record.SchemaUtil;
+import org.apache.drill.exec.record.VectorContainer;
+import org.apache.drill.exec.record.VectorWrapper;
+import org.apache.drill.exec.record.WritableBatch;
+import org.apache.drill.exec.record.selection.SelectionVector2;
+import org.apache.drill.exec.record.selection.SelectionVector4;
+import org.apache.drill.exec.testing.ControlsInjector;
+import org.apache.drill.exec.testing.ControlsInjectorFactory;
+import org.apache.drill.exec.vector.ValueVector;
+import org.apache.drill.exec.vector.complex.AbstractContainerVector;
+
+import com.google.common.collect.Lists;
+
+/**
+ * External sort batch: a sort batch which can spill to disk in
+ * order to operate within a defined memory footprint.
+ * <p>
+ * <h4>Basic Operation</h4>
+ * The operator has three key phases:
+ * <p>
+ * <ul>
+ * <li>The load phase in which batches are read from upstream.</li>
+ * <li>The merge phase in which spilled batches are combined to
+ * reduce the number of files below the configured limit. (Best
+ * practice is to configure the system to avoid this phase.)
+ * <li>The delivery phase in which batches are combined to produce
+ * the final output.</li>
+ * </ul>
+ * During the load phase:
+ * <p>
+ * <ul>
+ * <li>The incoming (upstream) operator provides a series of batches.</li>
+ * <li>This operator sorts each batch, and accumulates them in an in-memory
+ * buffer.</li>
+ * <li>If the in-memory buffer becomes too large, this operator selects
+ * a subset of the buffered batches to spill.</li>
+ * <li>Each spill set is merged to create a new, sorted collection of
+ * batches, and each is spilled to disk.</li>
+ * <li>To allow the use of multiple disk storage, each spill group is
written
+ * round-robin to a set of spill directories.</li>
+ * </ul>
+ * <p>
+ * During the sort/merge phase:
+ * <p>
+ * <ul>
+ * <li>When the input operator is complete, this operator merges the
accumulated
+ * batches (which may be all in memory or partially on disk), and returns
+ * them to the output (downstream) operator in chunks of no more than
+ * 32K records.</li>
+ * <li>The final merge must combine a collection of in-memory and spilled
+ * batches. Several limits apply to the maximum "width" of this merge. For
+ * example, we each open spill run consumes a file handle, and we may wish
--- End diff --
Fixed.
---
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 infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
