[
https://issues.apache.org/jira/browse/SPARK-34204?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Nick Hryhoriev updated SPARK-34204:
-----------------------------------
Description:
input_file_name() function damage applying projection to the physical plan of
the query.
if use this function and a new column, column-oriented formats like parquet
and orc put all columns to Physical plan.
While without it, only selected columns uploaded.
In my case, performance influence is x30.
{code:java}
import org.apache.spark.sql.SparkSession
import org.apache.spark.sql.functions._
object TestSize {
def main(args: Array[String]): Unit = {
implicit val spark: SparkSession = SparkSession.builder()
.master("local")
.config("spark.sql.shuffle.partitions", "5")
.getOrCreate()
import spark.implicits._
val quey1 = spark.read.parquet(
"s3a:/part-00040-a19f0d20-eab3-48ef-be5a-602c7f9a8e58.c000.gz.parquet"
)
.select($"app_id", $"idfa", input_file_name().as("fileName"))
.distinct()
.count()
val quey2 = spark.read.parquet( "s3a:/part-00040-a19f0d20-eab3-48ef-be5a-
602c7f9a8e58.c000.gz.parquet" )
.select($"app_id", $"idfa", input_file_name().as("fileName"))
.distinct()
.count()
Thread.sleep(10000000000L)
}
}
{code}
Query 1 has all columns in the physical plan, while query 2 only two.
was:
input_file_name() function damage applying projection to the physical plan of
the query.
if use this function and a new column, column-oriented formats like parquet
and orc put all columns to Physical plan.
While without it, only selected columns uploaded.
In my case, performance influence is x30.
{code:java}
import org.apache.spark.sql.SparkSession
import org.apache.spark.sql.functions._
object TestSize {
def main(args: Array[String]): Unit = {
implicit val spark: SparkSession = SparkSession.builder()
.master("local")
.config("spark.sql.shuffle.partitions", "5")
.getOrCreate()
import spark.implicits._
spark.read.parquet(
"s3a:/part-00040-a19f0d20-eab3-48ef-be5a-602c7f9a8e58.c000.gz.parquet"
)
.select($"app_id", $"idfa", input_file_name().as("fileName"))
.distinct()
.count()
Thread.sleep(10000000000L)
}
}
{code}
> input_file_name() func damage applying projection to physical plan of query.
> ----------------------------------------------------------------------------
>
> Key: SPARK-34204
> URL: https://issues.apache.org/jira/browse/SPARK-34204
> Project: Spark
> Issue Type: Bug
> Components: SQL
> Affects Versions: 2.4.7
> Reporter: Nick Hryhoriev
> Priority: Major
>
> input_file_name() function damage applying projection to the physical plan of
> the query.
> if use this function and a new column, column-oriented formats like parquet
> and orc put all columns to Physical plan.
> While without it, only selected columns uploaded.
> In my case, performance influence is x30.
> {code:java}
> import org.apache.spark.sql.SparkSession
> import org.apache.spark.sql.functions._
> object TestSize {
> def main(args: Array[String]): Unit = {
> implicit val spark: SparkSession = SparkSession.builder()
> .master("local")
> .config("spark.sql.shuffle.partitions", "5")
> .getOrCreate()
> import spark.implicits._
> val quey1 = spark.read.parquet(
> "s3a:/part-00040-a19f0d20-eab3-48ef-be5a-602c7f9a8e58.c000.gz.parquet"
> )
> .select($"app_id", $"idfa", input_file_name().as("fileName"))
> .distinct()
> .count()
> val quey2 = spark.read.parquet( "s3a:/part-00040-a19f0d20-eab3-48ef-be5a-
> 602c7f9a8e58.c000.gz.parquet" )
> .select($"app_id", $"idfa", input_file_name().as("fileName"))
> .distinct()
> .count()
> Thread.sleep(10000000000L)
> }
> }
> {code}
> Query 1 has all columns in the physical plan, while query 2 only two.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
