drill-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From paul-rogers <...@git.apache.org>
Subject [GitHub] drill pull request #846: DRILL-5544: Out of heap running CTAS against text d...
Date Tue, 30 May 2017 21:52:06 GMT
Github user paul-rogers commented on a diff in the pull request:

    https://github.com/apache/drill/pull/846#discussion_r119181468
  
    --- Diff: exec/java-exec/src/main/java/org/apache/parquet/hadoop/ParquetColumnChunkPageWriteStore.java
---
    @@ -0,0 +1,269 @@
    +/*
    + * 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.parquet.hadoop;
    +
    +import static org.apache.parquet.column.statistics.Statistics.getStatsBasedOnType;
    +
    +import java.io.Closeable;
    +import java.io.IOException;
    +import java.util.List;
    +import java.util.Map;
    +import java.util.Set;
    +
    +import com.google.common.collect.Lists;
    +import com.google.common.collect.Maps;
    +import com.google.common.collect.Sets;
    +import org.apache.drill.exec.store.parquet.ParquetDirectByteBufferAllocator;
    +import org.apache.parquet.bytes.BytesInput;
    +import org.apache.parquet.bytes.CapacityByteArrayOutputStream;
    +import org.apache.parquet.column.ColumnDescriptor;
    +import org.apache.parquet.column.Encoding;
    +import org.apache.parquet.column.page.DictionaryPage;
    +import org.apache.parquet.column.page.PageWriteStore;
    +import org.apache.parquet.column.page.PageWriter;
    +import org.apache.parquet.column.statistics.Statistics;
    +import org.apache.parquet.format.converter.ParquetMetadataConverter;
    +import org.apache.parquet.hadoop.CodecFactory.BytesCompressor;
    +import org.apache.parquet.io.ParquetEncodingException;
    +import org.apache.parquet.schema.MessageType;
    +import org.apache.parquet.bytes.ByteBufferAllocator;
    +import org.slf4j.Logger;
    +import org.slf4j.LoggerFactory;
    +
    +/**
    + * This is a copy of ColumnChunkPageWriteStore from parquet library except of OutputStream
that is used here.
    + * Using of CapacityByteArrayOutputStream allows to use different ByteBuffer allocators.
    --- End diff --
    
    Please explain, perhaps in the JIRA or in a code comment, the benefit of using a ByteBuffer
here. Here is my question. Data comes from value vectors, where it is stored as direct memory.
But, individual values have to be retrieved from the vector and converted to some other form:
perhaps an int, a double or however Parquet represents strings.
    
    Parquet will then work its magic on the values: compressing them, dictionary encoding
them, etc.
    
    Finally, the resulting buffers are written to disk. If I/O is synchronous, data comes
from a heap buffer. If async, then from direct.
    
    What is the benefit of introducing ByteBuffer allocators here? Are we doing that to use
direct memory? What is the advantage of copying Parquet's likely heap buffers into direct
memory for buffering prior to writing to a file?
    
    Said another way, why can't we just use the original Parquet version of this code?
    
    Some bit of explanation would be very helpful for us reviewers...


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

Mime
View raw message