org.apache.pig.piggybank.storage

Class MultiStorage

java.lang.Object

org.apache.pig.StoreFunc

org.apache.pig.piggybank.storage.MultiStorage

All Implemented Interfaces:

StoreFuncInterface

public class MultiStorage
extends StoreFunc

The UDF is useful for splitting the output data into a bunch of directories and files dynamically based on user specified key field in the output tuple. Sample usage: A = LOAD 'mydata' USING PigStorage() as (a, b, c); STORE A INTO '/my/home/output' USING MultiStorage('/my/home/output','0', 'bz2', '\\t'); Parameter details:- ========== /my/home/output (Required) : The DFS path where output directories and files will be created. 0 (Required) : Index of field whose values should be used to create directories and files( field 'a' in this case). 'bz2' (Optional) : The compression type. Default is 'none'. Supported types are:- 'none', 'gz' and 'bz2' '\\t' (Optional) : Output field separator. Let 'a1', 'a2' be the unique values of field 'a'. Then output may look like this /my/home/output/a1/a1-0000 /my/home/output/a1/a1-0001 /my/home/output/a1/a1-0002 ... /my/home/output/a2/a2-0000 /my/home/output/a2/a2-0001 /my/home/output/a2/a2-0002 The prefix '0000*' is the task-id of the mapper/reducer task executing this store. In case user does a GROUP BY on the field followed by MultiStorage(), then its imperative that all tuples for a particular group will go exactly to 1 reducer. So in the above case for e.g. there will be only 1 file each under 'a1' and 'a2' directories. If the output is compressed,then the sub directories and the output files will be having the extension. Say for example in the above case if bz2 is used one file will look like ;/my/home/output.bz2/a1.bz2/a1-0000.bz2

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	MultiStorage.MultiStorageOutputFormat

Constructor Summary

Constructors

Constructor and Description
MultiStorage(String parentPathStr, String splitFieldIndex)
MultiStorage(String parentPathStr, String splitFieldIndex, String compression)
MultiStorage(String parentPathStr, String splitFieldIndex, String compression, String fieldDel) Constructor

Method Summary

Methods

Modifier and Type	Method and Description
org.apache.hadoop.mapreduce.OutputFormat	getOutputFormat() Return the OutputFormat associated with StoreFunc.
void	prepareToWrite(org.apache.hadoop.mapreduce.RecordWriter writer) Initialize StoreFunc to write data.
void	putNext(Tuple tuple) Write a tuple to the data store.
void	setStoreLocation(String location, org.apache.hadoop.mapreduce.Job job) Communicate to the storer the location where the data needs to be stored.
Boolean	supportsParallelWriteToStoreLocation() DAG execution engines like Tez support optimizing union by writing to output location in parallel from tasks of different vertices.

Methods inherited from class org.apache.pig.StoreFunc

checkSchema, cleanupOnFailure, cleanupOnFailureImpl, cleanupOnSuccess, relToAbsPathForStoreLocation, setStoreFuncUDFContextSignature, warn

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

MultiStorage

public MultiStorage(String parentPathStr,
            String splitFieldIndex)

MultiStorage

public MultiStorage(String parentPathStr,
            String splitFieldIndex,
            String compression)

MultiStorage

public MultiStorage(String parentPathStr,
            String splitFieldIndex,
            String compression,
            String fieldDel)

Constructor

Parameters:

parentPathStr - Parent output dir path (this will be specified in store statement, so MultiStorage don't use this parameter in reality. However, we don't want to change the construct to break backward compatibility)

splitFieldIndex - key field index

compression - 'bz2', 'bz', 'gz' or 'none'

fieldDel - Output record field delimiter.

Method Detail

putNext

public void putNext(Tuple tuple)
             throws IOException

Description copied from class: StoreFunc

Write a tuple to the data store.

Specified by:

putNext in interface StoreFuncInterface

Specified by:

putNext in class StoreFunc

Parameters:

tuple - the tuple to store.

Throws:

IOException - if an exception occurs during the write

getOutputFormat

public org.apache.hadoop.mapreduce.OutputFormat getOutputFormat()
                                                         throws IOException

Description copied from class: StoreFunc

Return the OutputFormat associated with StoreFunc. This will be called on the front end during planning and on the backend during execution.

Specified by:

getOutputFormat in interface StoreFuncInterface

Specified by:

getOutputFormat in class StoreFunc

Returns:

the OutputFormat associated with StoreFunc

Throws:

IOException - if an exception occurs while constructing the OutputFormat

prepareToWrite

public void prepareToWrite(org.apache.hadoop.mapreduce.RecordWriter writer)
                    throws IOException

Description copied from class: StoreFunc

Initialize StoreFunc to write data. This will be called during execution on the backend before the call to putNext.

Specified by:

prepareToWrite in interface StoreFuncInterface

Specified by:

prepareToWrite in class StoreFunc

Parameters:

writer - RecordWriter to use.

Throws:

IOException - if an exception occurs during initialization

setStoreLocation

public void setStoreLocation(String location,
                    org.apache.hadoop.mapreduce.Job job)
                      throws IOException

Description copied from class: StoreFunc

Communicate to the storer the location where the data needs to be stored. The location string passed to the StoreFunc here is the return value of StoreFunc.relToAbsPathForStoreLocation(String, Path) This method will be called in the frontend and backend multiple times. Implementations should bear in mind that this method is called multiple times and should ensure there are no inconsistent side effects due to the multiple calls. StoreFunc.checkSchema(ResourceSchema) will be called before any call to StoreFunc.setStoreLocation(String, Job).

Specified by:

setStoreLocation in interface StoreFuncInterface

Specified by:

setStoreLocation in class StoreFunc

Parameters:

location - Location returned by StoreFunc.relToAbsPathForStoreLocation(String, Path)

job - The Job object

Throws:

IOException - if the location is not valid.

supportsParallelWriteToStoreLocation

public Boolean supportsParallelWriteToStoreLocation()

Description copied from class: StoreFunc

DAG execution engines like Tez support optimizing union by writing to output location in parallel from tasks of different vertices. Commit is called once all the vertices in the union are complete. This eliminates need to have a separate phase to read data output from previous phases, union them and write out again. Enabling the union optimization requires the OutputFormat to 1) Support creation of different part file names for tasks of different vertices. Conflicting filenames can create data corruption and loss. For eg: If task 0 of vertex1 and vertex2 both create filename as part-r-00000, then one of the files will be overwritten when promoting from temporary to final location leading to data loss. FileOutputFormat has mapreduce.output.basename config which enables naming files differently in different vertices. Classes extending FileOutputFormat and those prefixing file names with mapreduce.output.basename value will not encounter conflict. Cases like HBaseStorage which write to key value store and do not produce files also should not face any conflict. 2) Support calling of commit once at the end takes care of promoting temporary files of the different vertices into the final location. For eg: FileOutputFormat commit algorithm handles promoting of files produced by tasks of different vertices into final output location without issues if there is no file name conflict. In cases like HBaseStorage, the TableOutputCommitter does nothing on commit. If custom OutputFormat used by the StoreFunc does not support the above two criteria, then false should be returned. Union optimization will be disabled for the StoreFunc. Default implementation returns null and in that case planner falls back to PigConfiguration.PIG_TEZ_OPT_UNION_SUPPORTED_STOREFUNCS and PigConfiguration.PIG_TEZ_OPT_UNION_UNSUPPORTED_STOREFUNCS settings to determine if the StoreFunc supports it.

Overrides:

supportsParallelWriteToStoreLocation in class StoreFunc