代码之家 › 专栏 › 技术社区 › Etibu

使用谓词下推连接两个数据集

phoenix apache-spark-sql hbase apache-spark scala

4

Etibu · 技术社区 · 7 年前

我有一个我从一个 数据集 凤凰台

val dfToJoin = sparkSession.createDataset(rddToJoin)
val tableDf = sparkSession
  .read
  .option("table", "table")
  .option("zkURL", "localhost")
  .format("org.apache.phoenix.spark")
  .load()
val joinedDf = dfToJoin.join(tableDf, "columnToJoinOn")

当我执行它时,似乎加载了整个数据库表来进行连接。

也: 比桌子小,我不知道这是否重要。

编辑:基本上我想用通过spark创建的数据集连接我的Phoenix表,而不需要将整个表提取到executor中。

*Project [FEATURE#21, SEQUENCE_IDENTIFIER#22, TAX_NUMBER#23, 
         WINDOW_NUMBER#24, uniqueIdentifier#5, readLength#6]
 +- *SortMergeJoin [FEATURE#21], [feature#4], Inner
     :- *Sort [FEATURE#21 ASC NULLS FIRST], false, 0
     :  +- Exchange hashpartitioning(FEATURE#21, 200)
     :     +- *Filter isnotnull(FEATURE#21)
     :        +- *Scan PhoenixRelation(FEATURES,localhost,false) 

    [FEATURE#21,SEQUENCE_IDENTIFIER#22,TAX_NUMBER#23,WINDOW_NUMBER#24] 
    PushedFilters: [IsNotNull(FEATURE)], ReadSchema: 

    struct<FEATURE:int,SEQUENCE_IDENTIFIER:string,TAX_NUMBER:int,
    WINDOW_NUMBER:int>
   +- *Sort [feature#4 ASC NULLS FIRST], false, 0
  +- Exchange hashpartitioning(feature#4, 200)
     +- *Filter isnotnull(feature#4)
        +- *SerializeFromObject [assertnotnull(input[0, utils.CaseClasses$QueryFeature, true], top level Product input object).feature AS feature#4, staticinvoke(class org.apache.spark.unsafe.types.UTF8String, StringType, fromString, assertnotnull(input[0, utils.CaseClasses$QueryFeature, true], top level Product input object).uniqueIdentifier, true) AS uniqueIdentifier#5, assertnotnull(input[0, utils.CaseClasses$QueryFeature, true], top level Product input object).readLength AS readLength#6]
           +- Scan ExternalRDDScan[obj#3]

正如您所看到的,equals过滤器不包含在pushed filters列表中,因此很明显没有发生谓词下推。

1 回复 | 直到 7 年前

1

7

mrsrinivas 7 年前

不是一个执行者的整个表格 )

因为不是直接的 filter 在凤凰台df上,我们只看到 *Filter isnotnull(FEATURE#21)

正如您所提到的,当您对Phoenix表应用过滤器时,它的数据更少。你把过滤器推到phoenix表上 feature 按查找列 feature_ids

//This spread across workers  - fully distributed
val dfToJoin = sparkSession.createDataset(rddToJoin)

//This sits in driver - not distributed
val list_of_feature_ids = dfToJoin.dropDuplicates("feature")
  .select("feature")
  .map(r => r.getString(0))
  .collect
  .toList

//This spread across workers  - fully distributed
val tableDf = sparkSession
  .read
  .option("table", "table")
  .option("zkURL", "localhost")
  .format("org.apache.phoenix.spark")
  .load()
  .filter($"FEATURE".isin(list_of_feature_ids:_*)) //added filter

//This spread across workers  - fully distributed
val joinedDf = dfToJoin.join(tableDf, "columnToJoinOn")

joinedDf.explain()