Apache Spark User List

回复：Re: Is Spark SQL able to auto update partition stats like hive by setting hive.stats.autogather=true

_‹ Previous Topic Next Topic _›

Classic

List

Threaded

2 messages Options

疯狂的哈丘

回复：Re: Is Spark SQL able to auto update partition stats like hive by setting hive.stats.autogather=true

thx,but `hive.stats.autogather` is not work for sparkSQL.

----- 原始邮件 -----
发件人：Mich Talebzadeh <[hidden email]>
收件人：[hidden email]
抄送人：user <[hidden email]>
主题：Re: Is Spark SQL able to auto update partition stats like hive by setting hive.stats.autogather=true
日期：2020年12月19日 06点45分

Hi,

A fellow forum member kindly spotted a lousy error of mine, where a comma was missing at the line above the red line.

This appears to be accepted

spark = SparkSession.builder \
        .appName("app1") \
        .enableHiveSupport() \
        .getOrCreate()
# Hive settings
settings = [
    ("hive.exec.dynamic.partition", "true"),
    ("hive.exec.dynamic.partition.mode", "nonstrict"),
    ("spark.sql.orc.filterPushdown", "true"),
    ("hive.msck.path.validation", "ignore"),
    ("spark.sql.caseSensitive", "true"),
    ("spark.speculation", "false"),
    ("hive.metastore.authorization.storage.checks", "false"),
    ("hive.metastore.client.connect.retry.delay", "5s"),
    ("hive.metastore.client.socket.timeout", "1800s"),
    ("hive.metastore.connect.retries", "12"),
    ("hive.metastore.execute.setugi", "false"),
    ("hive.metastore.failure.retries", "12"),
    ("hive.metastore.schema.verification", "false"),
    ("hive.metastore.schema.verification.record.version", "false"),
    ("hive.metastore.server.max.threads", "100000"),
    ("hive.metastore.authorization.storage.checks", "/apps/hive/warehouse"),
    ("hive.stats.autogather", "true")
]
spark.sparkContext._conf.setAll(settings)

However, I have not tested it myself.

HTH

Mich

LinkedIn https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw

Disclaimer: Use it at your own risk. Any and all responsibility for any loss, damage or destruction of data or any other property which may arise from relying on this email's technical content is explicitly disclaimed. The author will in no case be liable for any monetary damages arising from such loss, damage or destruction.

On Fri, 18 Dec 2020 at 18:53, Mich Talebzadeh <[hidden email]> wrote:

I am afraid not supported for spark sql

see Automatic Statistics Collection For Better Query Performance | Qubole

I tried it as below
spark = SparkSession.builder \
        .appName("app1") \
        .enableHiveSupport() \
        .getOrCreate()
# Hive settings
settings = [
    ("hive.exec.dynamic.partition", "true"),
    ("hive.exec.dynamic.partition.mode", "nonstrict"),
    ("spark.sql.orc.filterPushdown", "true"),
    ("hive.msck.path.validation", "ignore"),
    ("spark.sql.caseSensitive", "true"),
    ("spark.speculation", "false"),
    ("hive.metastore.authorization.storage.checks", "false"),
    ("hive.metastore.client.connect.retry.delay", "5s"),
    ("hive.metastore.client.socket.timeout", "1800s"),
    ("hive.metastore.connect.retries", "12"),
    ("hive.metastore.execute.setugi", "false"),
    ("hive.metastore.failure.retries", "12"),
    ("hive.metastore.schema.verification", "false"),
    ("hive.metastore.schema.verification.record.version", "false"),
    ("hive.metastore.server.max.threads", "100000"),
    ("hive.metastore.authorization.storage.checks", "/apps/hive/warehouse")
    ("hive.stats.autogather", "true")
]
spark.sparkContext._conf.setAll(settings)
got this error

("hive.stats.autogather", "true")
TypeError: 'tuple' object is not callable

HTH

LinkedIn https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw

Disclaimer: Use it at your own risk. Any and all responsibility for any loss, damage or destruction of data or any other property which may arise from relying on this email's technical content is explicitly disclaimed. The author will in no case be liable for any monetary damages arising from such loss, damage or destruction.
On Fri, 18 Dec 2020 at 06:00, 疯狂的哈丘 <[hidden email]> wrote:
`spark.sql.statistics.size.autoUpdate.enabled` is only work for table stats update.But for partition stats,I can only update it with `ANALYZE TABLE tablename PARTITION(part) COMPUTE STATISTICS`.So is Spark SQL able to auto update partition stats like hive by setting hive.stats.autogather=true?

Mich Talebzadeh

Re: Re: Is Spark SQL able to auto update partition stats like hive by setting hive.stats.autogather=true

Ok if not working then you need to find a work around to update stats before

if (spark.sql(f"""SHOW TABLES IN {v.DB} like '{v.tableName}'""").count() == 1):
  spark.sql(f"""ANALYZE TABLE {v.fullyQualifiedTableName} compute statistics""")
  rows = spark.sql(f"""SELECT COUNT(1) FROM {v.fullyQualifiedTableName}""").collect()[0][0]
  print ("number of rows is ",rows)
else:
  print(f"\nTable {v.fullyQualifiedTableName} does not exist, creating table ")

HTH

LinkedIn https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw

On Sat, 19 Dec 2020 at 07:51, 疯狂的哈丘 <[hidden email]> wrote:

thx,but `hive.stats.autogather` is not work for sparkSQL.
----- 原始邮件 -----
发件人：Mich Talebzadeh <[hidden email]>
收件人：[hidden email]
抄送人：user <[hidden email]>
主题：Re: Is Spark SQL able to auto update partition stats like hive by setting hive.stats.autogather=true
日期：2020年12月19日 06点45分
Hi,

A fellow forum member kindly spotted a lousy error of mine, where a comma was missing at the line above the red line.

This appears to be accepted
spark = SparkSession.builder \
        .appName("app1") \
        .enableHiveSupport() \
        .getOrCreate()
# Hive settings
settings = [
    ("hive.exec.dynamic.partition", "true"),
    ("hive.exec.dynamic.partition.mode", "nonstrict"),
    ("spark.sql.orc.filterPushdown", "true"),
    ("hive.msck.path.validation", "ignore"),
    ("spark.sql.caseSensitive", "true"),
    ("spark.speculation", "false"),
    ("hive.metastore.authorization.storage.checks", "false"),
    ("hive.metastore.client.connect.retry.delay", "5s"),
    ("hive.metastore.client.socket.timeout", "1800s"),
    ("hive.metastore.connect.retries", "12"),
    ("hive.metastore.execute.setugi", "false"),
    ("hive.metastore.failure.retries", "12"),
    ("hive.metastore.schema.verification", "false"),
    ("hive.metastore.schema.verification.record.version", "false"),
    ("hive.metastore.server.max.threads", "100000"),
    ("hive.metastore.authorization.storage.checks", "/apps/hive/warehouse"),
    ("hive.stats.autogather", "true")
]
spark.sparkContext._conf.setAll(settings)
However, I have not tested it myself.

HTH

Mich

LinkedIn https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw

Disclaimer: Use it at your own risk. Any and all responsibility for any loss, damage or destruction of data or any other property which may arise from relying on this email's technical content is explicitly disclaimed. The author will in no case be liable for any monetary damages arising from such loss, damage or destruction.
On Fri, 18 Dec 2020 at 18:53, Mich Talebzadeh <[hidden email]> wrote:
I am afraid not supported for spark sql

see Automatic Statistics Collection For Better Query Performance | Qubole

I tried it as below
spark = SparkSession.builder \
        .appName("app1") \
        .enableHiveSupport() \
        .getOrCreate()
# Hive settings
settings = [
    ("hive.exec.dynamic.partition", "true"),
    ("hive.exec.dynamic.partition.mode", "nonstrict"),
    ("spark.sql.orc.filterPushdown", "true"),
    ("hive.msck.path.validation", "ignore"),
    ("spark.sql.caseSensitive", "true"),
    ("spark.speculation", "false"),
    ("hive.metastore.authorization.storage.checks", "false"),
    ("hive.metastore.client.connect.retry.delay", "5s"),
    ("hive.metastore.client.socket.timeout", "1800s"),
    ("hive.metastore.connect.retries", "12"),
    ("hive.metastore.execute.setugi", "false"),
    ("hive.metastore.failure.retries", "12"),
    ("hive.metastore.schema.verification", "false"),
    ("hive.metastore.schema.verification.record.version", "false"),
    ("hive.metastore.server.max.threads", "100000"),
    ("hive.metastore.authorization.storage.checks", "/apps/hive/warehouse")
    ("hive.stats.autogather", "true")
]
spark.sparkContext._conf.setAll(settings)
got this error

("hive.stats.autogather", "true")
TypeError: 'tuple' object is not callable

HTH

LinkedIn https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw

Disclaimer: Use it at your own risk. Any and all responsibility for any loss, damage or destruction of data or any other property which may arise from relying on this email's technical content is explicitly disclaimed. The author will in no case be liable for any monetary damages arising from such loss, damage or destruction.
On Fri, 18 Dec 2020 at 06:00, 疯狂的哈丘 <[hidden email]> wrote:
`spark.sql.statistics.size.autoUpdate.enabled` is only work for table stats update.But for partition stats,I can only update it with `ANALYZE TABLE tablename PARTITION(part) COMPUTE STATISTICS`.So is Spark SQL able to auto update partition stats like hive by setting hive.stats.autogather=true?