使用Binary Memtable将大量数据导入Cassandra

在这篇《谈谈Cassandra的客户端》文章中，我们谈到了如何使用Thrift API以及更加高级的封装（Hector）如果将数据导入到到Cassandra中，但是在导入大量数据的时候这会遇到很多的问题，比如插入失败，超时等等问题。为了解决这个问题，我们可以尝试使用Binary Memtable。

在Cassandra的wiki上，对Binary Memtable的描述如下：

Binary Memtable is the name of Cassandra's bulk-load interface. It avoids several kinds of overhead associated with the normal Thrift API:

• Converting to Thrift from the internal structures and back
• Routing (copying) from a coordinator node to the replica nodes
• Writing to the commitlog
• Serializing the internal structures to on-disk format

The tradeoff you make is that it is considerably less convenient to use than Thrift:

• You must use the StorageProxy API, only available as Java code

• You must pre-serialize the rows yourself

• The rows you send are not live for querying until a flush occurs (either normally because the Binary Memtable fills up, or because you request one with nodetool)

• You must write an entire row at once

具体的实现我们可以参考https://svn.apache.org/repos/asf/cassandra/trunk/contrib/bmt_example/.但是这个示例结合了Hadoop，如何没有这个环境，也可以参考我的测试代码：

package com.alibaba.dw.thrift.client;

import java.io.IOException;
import java.io.UnsupportedEncodingException;

import org.apache.cassandra.config.DatabaseDescriptor;
import org.apache.cassandra.db.Column;
import org.apache.cassandra.db.ColumnFamily;
import org.apache.cassandra.db.RowMutation;
import org.apache.cassandra.db.filter.QueryPath;
import org.apache.cassandra.io.util.DataOutputBuffer;
import java.net.InetAddress;

import org.apache.cassandra.net.Message;
import org.apache.cassandra.net.MessagingService;
import org.apache.cassandra.service.StorageService;

/**
 * TODO Comment of BinaryMemtableTest
 *
 * @author aaron.guop
 */
public class BinaryMemtableTest {

    /**
     * @param args
     * @throws IOException
     * @throws InterruptedException
     */
    public static void main(String[] args) throws IOException, InterruptedException {
        System.setProperty("storage-config", "D:\\apache-cassandra-0.6.1\\conf");

        StorageService.instance.initClient();

        while (StorageService.instance.getNaturalEndpoints("Keyspace1", "bmt").isEmpty()) {
            Thread.sleep(1 * 1000);
        }


        doInsert();

        StorageService.instance.stopClient();
    }

    /**
     * @throws UnsupportedEncodingException
     */
    private static void doInsert() throws UnsupportedEncodingException {
        String keyspace = "Keyspace1";
        String cfName = "Member";
        String memberID = "bmt";

        /* Create a column family */
        ColumnFamily columnFamily = ColumnFamily.create(keyspace, cfName);

        //while (values.hasNext()) {
        String SuperColumnName = "SuperColumnName";
        String ColumnName = "ColumnName";
        String ColumnValue = "ColumnValue";
        long timestamp = 0;
        columnFamily.addColumn(new QueryPath(cfName, SuperColumnName.getBytes("UTF-8"), ColumnName
                .getBytes("UTF-8")), ColumnValue.getBytes(), timestamp);
        //}

        /* Get serialized message to send to cluster */
        Message message = createMessage(keyspace, memberID, cfName, columnFamily);
        for (InetAddress endpoint : StorageService.instance.getNaturalEndpoints(keyspace, memberID)) {
            /* Send message to end point */
            MessagingService.instance.sendOneWay(message, endpoint);
            System.out.println("Send message to " + endpoint.toString());
        }
    }

    public static Message createMessage(String Keyspace, String Key, String CFName,
                                        ColumnFamily columnFamile) {
        DataOutputBuffer bufOut = new DataOutputBuffer();
        Column column;

        /*
         * Get the first column family from list, this is just to get past
         * validation
         */
        ColumnFamily baseColumnFamily = new ColumnFamily(CFName, "Standard", DatabaseDescriptor
                .getComparator(Keyspace, CFName), DatabaseDescriptor.getSubComparator(Keyspace,
                CFName));

        bufOut.reset();
        try {
            ColumnFamily.serializer().serializeWithIndexes(columnFamile, bufOut);
            byte[] data = new byte[bufOut.getLength()];
            System.arraycopy(bufOut.getData(), 0, data, 0, bufOut.getLength());

            column = new Column(columnFamile.name().getBytes("UTF-8"), data, 0, false);
            baseColumnFamily.addColumn(column);
        } catch (IOException e) {
            throw new RuntimeException(e);
        }

        RowMutation rm = new RowMutation(Keyspace, Key);
        rm.add(baseColumnFamily);

        try {
            /* Make message */
            return rm.makeRowMutationMessage(StorageService.Verb.BINARY);
        } catch (IOException e) {
            throw new RuntimeException(e);
        }
    }

}

在运行这个代码之前，我们还必须注意以下事项：

1 运行这个程序的机器不能运行Cassandra Server。

2 在storage-conf.xml的seed中指定其他Cassandra Server的地址。

3 使用nodetool的getcachecapacity得到当前的压缩设置，记录下来，供完成导入后恢复使用。

4 使用nodetool的setcachecapacity设置当前的压缩设置为0，即取消压缩功能。

5 运行导入程序。

6 使用nodetool的flush。

7 使用nodetool的setcompactionthreshold设置回之前在第三步获得的值。 

经过这些操作后，数据就导入到我们的Cassandra中了。

 

更多关于Cassandra的文章：http://www.cnblogs.com/gpcuster/tag/Cassandra/