[jira] [Commented] (SPARK-41497) Accumulator undercounting in the case of retry task with rdd cache
[
https://issues.apache.org/jira/browse/SPARK-41497?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17696526#comment-17696526
]
Apache Spark commented on SPARK-41497:
--
User 'ivoson' has created a pull request for this issue:
https://github.com/apache/spark/pull/40281
> Accumulator undercounting in the case of retry task with rdd cache
> --
>
> Key: SPARK-41497
> URL: https://issues.apache.org/jira/browse/SPARK-41497
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
>Affects Versions: 2.4.8, 3.0.3, 3.1.3, 3.2.2, 3.3.1
>Reporter: wuyi
>Assignee: Tengfei Huang
>Priority: Major
> Fix For: 3.5.0
>
>
> Accumulator could be undercounted when the retried task has rdd cache. See
> the example below and you could also find the completed and reproducible
> example at
> [https://github.com/apache/spark/compare/master...Ngone51:spark:fix-acc]
>
> {code:scala}
> test("SPARK-XXX") {
> // Set up a cluster with 2 executors
> val conf = new SparkConf()
> .setMaster("local-cluster[2, 1,
> 1024]").setAppName("TaskSchedulerImplSuite")
> sc = new SparkContext(conf)
> // Set up a custom task scheduler. The scheduler will fail the first task
> attempt of the job
> // submitted below. In particular, the failed first attempt task would
> success on computation
> // (accumulator accounting, result caching) but only fail to report its
> success status due
> // to the concurrent executor lost. The second task attempt would success.
> taskScheduler = setupSchedulerWithCustomStatusUpdate(sc)
> val myAcc = sc.longAccumulator("myAcc")
> // Initiate a rdd with only one partition so there's only one task and
> specify the storage level
> // with MEMORY_ONLY_2 so that the rdd result will be cached on both two
> executors.
> val rdd = sc.parallelize(0 until 10, 1).mapPartitions { iter =>
> myAcc.add(100)
> iter.map(x => x + 1)
> }.persist(StorageLevel.MEMORY_ONLY_2)
> // This will pass since the second task attempt will succeed
> assert(rdd.count() === 10)
> // This will fail due to `myAcc.add(100)` won't be executed during the
> second task attempt's
> // execution. Because the second task attempt will load the rdd cache
> directly instead of
> // executing the task function so `myAcc.add(100)` is skipped.
> assert(myAcc.value === 100)
> } {code}
>
> We could also hit this issue with decommission even if the rdd only has one
> copy. For example, decommission could migrate the rdd cache block to another
> executor (the result is actually the same with 2 copies) and the
> decommissioned executor lost before the task reports its success status to
> the driver.
>
> And the issue is a bit more complicated than expected to fix. I have tried to
> give some fixes but all of them are not ideal:
> Option 1: Clean up any rdd cache related to the failed task: in practice,
> this option can already fix the issue in most cases. However, theoretically,
> rdd cache could be reported to the driver right after the driver cleans up
> the failed task's caches due to asynchronous communication. So this option
> can’t resolve the issue thoroughly;
> Option 2: Disallow rdd cache reuse across the task attempts for the same
> task: this option can 100% fix the issue. The problem is this way can also
> affect the case where rdd cache can be reused across the attempts (e.g., when
> there is no accumulator operation in the task), which can have perf
> regression;
> Option 3: Introduce accumulator cache: first, this requires a new framework
> for supporting accumulator cache; second, the driver should improve its logic
> to distinguish whether the accumulator cache value should be reported to the
> user to avoid overcounting. For example, in the case of task retry, the value
> should be reported. However, in the case of rdd cache reuse, the value
> shouldn’t be reported (should it?);
> Option 4: Do task success validation when a task trying to load the rdd
> cache: this way defines a rdd cache is only valid/accessible if the task has
> succeeded. This way could be either overkill or a bit complex (because
> currently Spark would clean up the task state once it’s finished. So we need
> to maintain a structure to know if task once succeeded or not. )
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
-
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-41497) Accumulator undercounting in the case of retry task with rdd cache
[
https://issues.apache.org/jira/browse/SPARK-41497?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17655807#comment-17655807
]
Apache Spark commented on SPARK-41497:
--
User 'ivoson' has created a pull request for this issue:
https://github.com/apache/spark/pull/39459
> Accumulator undercounting in the case of retry task with rdd cache
> --
>
> Key: SPARK-41497
> URL: https://issues.apache.org/jira/browse/SPARK-41497
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
>Affects Versions: 2.4.8, 3.0.3, 3.1.3, 3.2.2, 3.3.1
>Reporter: wuyi
>Priority: Major
>
> Accumulator could be undercounted when the retried task has rdd cache. See
> the example below and you could also find the completed and reproducible
> example at
> [https://github.com/apache/spark/compare/master...Ngone51:spark:fix-acc]
>
> {code:scala}
> test("SPARK-XXX") {
> // Set up a cluster with 2 executors
> val conf = new SparkConf()
> .setMaster("local-cluster[2, 1,
> 1024]").setAppName("TaskSchedulerImplSuite")
> sc = new SparkContext(conf)
> // Set up a custom task scheduler. The scheduler will fail the first task
> attempt of the job
> // submitted below. In particular, the failed first attempt task would
> success on computation
> // (accumulator accounting, result caching) but only fail to report its
> success status due
> // to the concurrent executor lost. The second task attempt would success.
> taskScheduler = setupSchedulerWithCustomStatusUpdate(sc)
> val myAcc = sc.longAccumulator("myAcc")
> // Initiate a rdd with only one partition so there's only one task and
> specify the storage level
> // with MEMORY_ONLY_2 so that the rdd result will be cached on both two
> executors.
> val rdd = sc.parallelize(0 until 10, 1).mapPartitions { iter =>
> myAcc.add(100)
> iter.map(x => x + 1)
> }.persist(StorageLevel.MEMORY_ONLY_2)
> // This will pass since the second task attempt will succeed
> assert(rdd.count() === 10)
> // This will fail due to `myAcc.add(100)` won't be executed during the
> second task attempt's
> // execution. Because the second task attempt will load the rdd cache
> directly instead of
> // executing the task function so `myAcc.add(100)` is skipped.
> assert(myAcc.value === 100)
> } {code}
>
> We could also hit this issue with decommission even if the rdd only has one
> copy. For example, decommission could migrate the rdd cache block to another
> executor (the result is actually the same with 2 copies) and the
> decommissioned executor lost before the task reports its success status to
> the driver.
>
> And the issue is a bit more complicated than expected to fix. I have tried to
> give some fixes but all of them are not ideal:
> Option 1: Clean up any rdd cache related to the failed task: in practice,
> this option can already fix the issue in most cases. However, theoretically,
> rdd cache could be reported to the driver right after the driver cleans up
> the failed task's caches due to asynchronous communication. So this option
> can’t resolve the issue thoroughly;
> Option 2: Disallow rdd cache reuse across the task attempts for the same
> task: this option can 100% fix the issue. The problem is this way can also
> affect the case where rdd cache can be reused across the attempts (e.g., when
> there is no accumulator operation in the task), which can have perf
> regression;
> Option 3: Introduce accumulator cache: first, this requires a new framework
> for supporting accumulator cache; second, the driver should improve its logic
> to distinguish whether the accumulator cache value should be reported to the
> user to avoid overcounting. For example, in the case of task retry, the value
> should be reported. However, in the case of rdd cache reuse, the value
> shouldn’t be reported (should it?);
> Option 4: Do task success validation when a task trying to load the rdd
> cache: this way defines a rdd cache is only valid/accessible if the task has
> succeeded. This way could be either overkill or a bit complex (because
> currently Spark would clean up the task state once it’s finished. So we need
> to maintain a structure to know if task once succeeded or not. )
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
-
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-41497) Accumulator undercounting in the case of retry task with rdd cache
[
https://issues.apache.org/jira/browse/SPARK-41497?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17655806#comment-17655806
]
Apache Spark commented on SPARK-41497:
--
User 'ivoson' has created a pull request for this issue:
https://github.com/apache/spark/pull/39459
> Accumulator undercounting in the case of retry task with rdd cache
> --
>
> Key: SPARK-41497
> URL: https://issues.apache.org/jira/browse/SPARK-41497
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
>Affects Versions: 2.4.8, 3.0.3, 3.1.3, 3.2.2, 3.3.1
>Reporter: wuyi
>Priority: Major
>
> Accumulator could be undercounted when the retried task has rdd cache. See
> the example below and you could also find the completed and reproducible
> example at
> [https://github.com/apache/spark/compare/master...Ngone51:spark:fix-acc]
>
> {code:scala}
> test("SPARK-XXX") {
> // Set up a cluster with 2 executors
> val conf = new SparkConf()
> .setMaster("local-cluster[2, 1,
> 1024]").setAppName("TaskSchedulerImplSuite")
> sc = new SparkContext(conf)
> // Set up a custom task scheduler. The scheduler will fail the first task
> attempt of the job
> // submitted below. In particular, the failed first attempt task would
> success on computation
> // (accumulator accounting, result caching) but only fail to report its
> success status due
> // to the concurrent executor lost. The second task attempt would success.
> taskScheduler = setupSchedulerWithCustomStatusUpdate(sc)
> val myAcc = sc.longAccumulator("myAcc")
> // Initiate a rdd with only one partition so there's only one task and
> specify the storage level
> // with MEMORY_ONLY_2 so that the rdd result will be cached on both two
> executors.
> val rdd = sc.parallelize(0 until 10, 1).mapPartitions { iter =>
> myAcc.add(100)
> iter.map(x => x + 1)
> }.persist(StorageLevel.MEMORY_ONLY_2)
> // This will pass since the second task attempt will succeed
> assert(rdd.count() === 10)
> // This will fail due to `myAcc.add(100)` won't be executed during the
> second task attempt's
> // execution. Because the second task attempt will load the rdd cache
> directly instead of
> // executing the task function so `myAcc.add(100)` is skipped.
> assert(myAcc.value === 100)
> } {code}
>
> We could also hit this issue with decommission even if the rdd only has one
> copy. For example, decommission could migrate the rdd cache block to another
> executor (the result is actually the same with 2 copies) and the
> decommissioned executor lost before the task reports its success status to
> the driver.
>
> And the issue is a bit more complicated than expected to fix. I have tried to
> give some fixes but all of them are not ideal:
> Option 1: Clean up any rdd cache related to the failed task: in practice,
> this option can already fix the issue in most cases. However, theoretically,
> rdd cache could be reported to the driver right after the driver cleans up
> the failed task's caches due to asynchronous communication. So this option
> can’t resolve the issue thoroughly;
> Option 2: Disallow rdd cache reuse across the task attempts for the same
> task: this option can 100% fix the issue. The problem is this way can also
> affect the case where rdd cache can be reused across the attempts (e.g., when
> there is no accumulator operation in the task), which can have perf
> regression;
> Option 3: Introduce accumulator cache: first, this requires a new framework
> for supporting accumulator cache; second, the driver should improve its logic
> to distinguish whether the accumulator cache value should be reported to the
> user to avoid overcounting. For example, in the case of task retry, the value
> should be reported. However, in the case of rdd cache reuse, the value
> shouldn’t be reported (should it?);
> Option 4: Do task success validation when a task trying to load the rdd
> cache: this way defines a rdd cache is only valid/accessible if the task has
> succeeded. This way could be either overkill or a bit complex (because
> currently Spark would clean up the task state once it’s finished. So we need
> to maintain a structure to know if task once succeeded or not. )
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
-
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-41497) Accumulator undercounting in the case of retry task with rdd cache
[
https://issues.apache.org/jira/browse/SPARK-41497?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17654590#comment-17654590
]
Mridul Muralidharan commented on SPARK-41497:
-
Sounds good [~Ngone51], thanks !
> Accumulator undercounting in the case of retry task with rdd cache
> --
>
> Key: SPARK-41497
> URL: https://issues.apache.org/jira/browse/SPARK-41497
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
>Affects Versions: 2.4.8, 3.0.3, 3.1.3, 3.2.2, 3.3.1
>Reporter: wuyi
>Priority: Major
>
> Accumulator could be undercounted when the retried task has rdd cache. See
> the example below and you could also find the completed and reproducible
> example at
> [https://github.com/apache/spark/compare/master...Ngone51:spark:fix-acc]
>
> {code:scala}
> test("SPARK-XXX") {
> // Set up a cluster with 2 executors
> val conf = new SparkConf()
> .setMaster("local-cluster[2, 1,
> 1024]").setAppName("TaskSchedulerImplSuite")
> sc = new SparkContext(conf)
> // Set up a custom task scheduler. The scheduler will fail the first task
> attempt of the job
> // submitted below. In particular, the failed first attempt task would
> success on computation
> // (accumulator accounting, result caching) but only fail to report its
> success status due
> // to the concurrent executor lost. The second task attempt would success.
> taskScheduler = setupSchedulerWithCustomStatusUpdate(sc)
> val myAcc = sc.longAccumulator("myAcc")
> // Initiate a rdd with only one partition so there's only one task and
> specify the storage level
> // with MEMORY_ONLY_2 so that the rdd result will be cached on both two
> executors.
> val rdd = sc.parallelize(0 until 10, 1).mapPartitions { iter =>
> myAcc.add(100)
> iter.map(x => x + 1)
> }.persist(StorageLevel.MEMORY_ONLY_2)
> // This will pass since the second task attempt will succeed
> assert(rdd.count() === 10)
> // This will fail due to `myAcc.add(100)` won't be executed during the
> second task attempt's
> // execution. Because the second task attempt will load the rdd cache
> directly instead of
> // executing the task function so `myAcc.add(100)` is skipped.
> assert(myAcc.value === 100)
> } {code}
>
> We could also hit this issue with decommission even if the rdd only has one
> copy. For example, decommission could migrate the rdd cache block to another
> executor (the result is actually the same with 2 copies) and the
> decommissioned executor lost before the task reports its success status to
> the driver.
>
> And the issue is a bit more complicated than expected to fix. I have tried to
> give some fixes but all of them are not ideal:
> Option 1: Clean up any rdd cache related to the failed task: in practice,
> this option can already fix the issue in most cases. However, theoretically,
> rdd cache could be reported to the driver right after the driver cleans up
> the failed task's caches due to asynchronous communication. So this option
> can’t resolve the issue thoroughly;
> Option 2: Disallow rdd cache reuse across the task attempts for the same
> task: this option can 100% fix the issue. The problem is this way can also
> affect the case where rdd cache can be reused across the attempts (e.g., when
> there is no accumulator operation in the task), which can have perf
> regression;
> Option 3: Introduce accumulator cache: first, this requires a new framework
> for supporting accumulator cache; second, the driver should improve its logic
> to distinguish whether the accumulator cache value should be reported to the
> user to avoid overcounting. For example, in the case of task retry, the value
> should be reported. However, in the case of rdd cache reuse, the value
> shouldn’t be reported (should it?);
> Option 4: Do task success validation when a task trying to load the rdd
> cache: this way defines a rdd cache is only valid/accessible if the task has
> succeeded. This way could be either overkill or a bit complex (because
> currently Spark would clean up the task state once it’s finished. So we need
> to maintain a structure to know if task once succeeded or not. )
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
-
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-41497) Accumulator undercounting in the case of retry task with rdd cache
[
https://issues.apache.org/jira/browse/SPARK-41497?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17653781#comment-17653781
]
wuyi commented on SPARK-41497:
--
> If I am not wrong, SQL makes very heavy use of accumulators, and so most
> stages will end up having them anyway - right ?
Right.
> I would expect this scenario (even without accumulator) to be fairly low
> frequency enough that the cost of extra recomputation might be fine.
Agree. So shall we proceed with the improved Option 4 that was proposed by you
[~mridulm80] ? [~ivoson] can help with the fix.
> Accumulator undercounting in the case of retry task with rdd cache
> --
>
> Key: SPARK-41497
> URL: https://issues.apache.org/jira/browse/SPARK-41497
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
>Affects Versions: 2.4.8, 3.0.3, 3.1.3, 3.2.2, 3.3.1
>Reporter: wuyi
>Priority: Major
>
> Accumulator could be undercounted when the retried task has rdd cache. See
> the example below and you could also find the completed and reproducible
> example at
> [https://github.com/apache/spark/compare/master...Ngone51:spark:fix-acc]
>
> {code:scala}
> test("SPARK-XXX") {
> // Set up a cluster with 2 executors
> val conf = new SparkConf()
> .setMaster("local-cluster[2, 1,
> 1024]").setAppName("TaskSchedulerImplSuite")
> sc = new SparkContext(conf)
> // Set up a custom task scheduler. The scheduler will fail the first task
> attempt of the job
> // submitted below. In particular, the failed first attempt task would
> success on computation
> // (accumulator accounting, result caching) but only fail to report its
> success status due
> // to the concurrent executor lost. The second task attempt would success.
> taskScheduler = setupSchedulerWithCustomStatusUpdate(sc)
> val myAcc = sc.longAccumulator("myAcc")
> // Initiate a rdd with only one partition so there's only one task and
> specify the storage level
> // with MEMORY_ONLY_2 so that the rdd result will be cached on both two
> executors.
> val rdd = sc.parallelize(0 until 10, 1).mapPartitions { iter =>
> myAcc.add(100)
> iter.map(x => x + 1)
> }.persist(StorageLevel.MEMORY_ONLY_2)
> // This will pass since the second task attempt will succeed
> assert(rdd.count() === 10)
> // This will fail due to `myAcc.add(100)` won't be executed during the
> second task attempt's
> // execution. Because the second task attempt will load the rdd cache
> directly instead of
> // executing the task function so `myAcc.add(100)` is skipped.
> assert(myAcc.value === 100)
> } {code}
>
> We could also hit this issue with decommission even if the rdd only has one
> copy. For example, decommission could migrate the rdd cache block to another
> executor (the result is actually the same with 2 copies) and the
> decommissioned executor lost before the task reports its success status to
> the driver.
>
> And the issue is a bit more complicated than expected to fix. I have tried to
> give some fixes but all of them are not ideal:
> Option 1: Clean up any rdd cache related to the failed task: in practice,
> this option can already fix the issue in most cases. However, theoretically,
> rdd cache could be reported to the driver right after the driver cleans up
> the failed task's caches due to asynchronous communication. So this option
> can’t resolve the issue thoroughly;
> Option 2: Disallow rdd cache reuse across the task attempts for the same
> task: this option can 100% fix the issue. The problem is this way can also
> affect the case where rdd cache can be reused across the attempts (e.g., when
> there is no accumulator operation in the task), which can have perf
> regression;
> Option 3: Introduce accumulator cache: first, this requires a new framework
> for supporting accumulator cache; second, the driver should improve its logic
> to distinguish whether the accumulator cache value should be reported to the
> user to avoid overcounting. For example, in the case of task retry, the value
> should be reported. However, in the case of rdd cache reuse, the value
> shouldn’t be reported (should it?);
> Option 4: Do task success validation when a task trying to load the rdd
> cache: this way defines a rdd cache is only valid/accessible if the task has
> succeeded. This way could be either overkill or a bit complex (because
> currently Spark would clean up the task state once it’s finished. So we need
> to maintain a structure to know if task once succeeded or not. )
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
-
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-
[jira] [Commented] (SPARK-41497) Accumulator undercounting in the case of retry task with rdd cache
[
https://issues.apache.org/jira/browse/SPARK-41497?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17653648#comment-17653648
]
Juliusz Sompolski commented on SPARK-41497:
---
Note that this issue leads to a correctness issue in Delta Merge, because it
depends on a SetAccumulator as a side output channel for gathering files that
need to be rewritten by the Merge:
https://github.com/delta-io/delta/blob/master/core/src/main/scala/org/apache/spark/sql/delta/commands/MergeIntoCommand.scala#L445-L449
Missing some files there can result in duplicate records being inserted instead
of existing records being updated.
> Accumulator undercounting in the case of retry task with rdd cache
> --
>
> Key: SPARK-41497
> URL: https://issues.apache.org/jira/browse/SPARK-41497
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
>Affects Versions: 2.4.8, 3.0.3, 3.1.3, 3.2.2, 3.3.1
>Reporter: wuyi
>Priority: Major
>
> Accumulator could be undercounted when the retried task has rdd cache. See
> the example below and you could also find the completed and reproducible
> example at
> [https://github.com/apache/spark/compare/master...Ngone51:spark:fix-acc]
>
> {code:scala}
> test("SPARK-XXX") {
> // Set up a cluster with 2 executors
> val conf = new SparkConf()
> .setMaster("local-cluster[2, 1,
> 1024]").setAppName("TaskSchedulerImplSuite")
> sc = new SparkContext(conf)
> // Set up a custom task scheduler. The scheduler will fail the first task
> attempt of the job
> // submitted below. In particular, the failed first attempt task would
> success on computation
> // (accumulator accounting, result caching) but only fail to report its
> success status due
> // to the concurrent executor lost. The second task attempt would success.
> taskScheduler = setupSchedulerWithCustomStatusUpdate(sc)
> val myAcc = sc.longAccumulator("myAcc")
> // Initiate a rdd with only one partition so there's only one task and
> specify the storage level
> // with MEMORY_ONLY_2 so that the rdd result will be cached on both two
> executors.
> val rdd = sc.parallelize(0 until 10, 1).mapPartitions { iter =>
> myAcc.add(100)
> iter.map(x => x + 1)
> }.persist(StorageLevel.MEMORY_ONLY_2)
> // This will pass since the second task attempt will succeed
> assert(rdd.count() === 10)
> // This will fail due to `myAcc.add(100)` won't be executed during the
> second task attempt's
> // execution. Because the second task attempt will load the rdd cache
> directly instead of
> // executing the task function so `myAcc.add(100)` is skipped.
> assert(myAcc.value === 100)
> } {code}
>
> We could also hit this issue with decommission even if the rdd only has one
> copy. For example, decommission could migrate the rdd cache block to another
> executor (the result is actually the same with 2 copies) and the
> decommissioned executor lost before the task reports its success status to
> the driver.
>
> And the issue is a bit more complicated than expected to fix. I have tried to
> give some fixes but all of them are not ideal:
> Option 1: Clean up any rdd cache related to the failed task: in practice,
> this option can already fix the issue in most cases. However, theoretically,
> rdd cache could be reported to the driver right after the driver cleans up
> the failed task's caches due to asynchronous communication. So this option
> can’t resolve the issue thoroughly;
> Option 2: Disallow rdd cache reuse across the task attempts for the same
> task: this option can 100% fix the issue. The problem is this way can also
> affect the case where rdd cache can be reused across the attempts (e.g., when
> there is no accumulator operation in the task), which can have perf
> regression;
> Option 3: Introduce accumulator cache: first, this requires a new framework
> for supporting accumulator cache; second, the driver should improve its logic
> to distinguish whether the accumulator cache value should be reported to the
> user to avoid overcounting. For example, in the case of task retry, the value
> should be reported. However, in the case of rdd cache reuse, the value
> shouldn’t be reported (should it?);
> Option 4: Do task success validation when a task trying to load the rdd
> cache: this way defines a rdd cache is only valid/accessible if the task has
> succeeded. This way could be either overkill or a bit complex (because
> currently Spark would clean up the task state once it’s finished. So we need
> to maintain a structure to know if task once succeeded or not. )
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
-
To unsubscribe, e-mail: issues-unsubscr...@spa
[jira] [Commented] (SPARK-41497) Accumulator undercounting in the case of retry task with rdd cache
[
https://issues.apache.org/jira/browse/SPARK-41497?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17646972#comment-17646972
]
Mridul Muralidharan commented on SPARK-41497:
-
[~Ngone51] Agree, that is what I was not sure of.
I would expect this scenario (even without accumulator) to be fairly low
frequency enough that the cost of extra recomputation might be fine.
> Accumulator undercounting in the case of retry task with rdd cache
> --
>
> Key: SPARK-41497
> URL: https://issues.apache.org/jira/browse/SPARK-41497
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
>Affects Versions: 2.4.8, 3.0.3, 3.1.3, 3.2.2, 3.3.1
>Reporter: wuyi
>Priority: Major
>
> Accumulator could be undercounted when the retried task has rdd cache. See
> the example below and you could also find the completed and reproducible
> example at
> [https://github.com/apache/spark/compare/master...Ngone51:spark:fix-acc]
>
> {code:scala}
> test("SPARK-XXX") {
> // Set up a cluster with 2 executors
> val conf = new SparkConf()
> .setMaster("local-cluster[2, 1,
> 1024]").setAppName("TaskSchedulerImplSuite")
> sc = new SparkContext(conf)
> // Set up a custom task scheduler. The scheduler will fail the first task
> attempt of the job
> // submitted below. In particular, the failed first attempt task would
> success on computation
> // (accumulator accounting, result caching) but only fail to report its
> success status due
> // to the concurrent executor lost. The second task attempt would success.
> taskScheduler = setupSchedulerWithCustomStatusUpdate(sc)
> val myAcc = sc.longAccumulator("myAcc")
> // Initiate a rdd with only one partition so there's only one task and
> specify the storage level
> // with MEMORY_ONLY_2 so that the rdd result will be cached on both two
> executors.
> val rdd = sc.parallelize(0 until 10, 1).mapPartitions { iter =>
> myAcc.add(100)
> iter.map(x => x + 1)
> }.persist(StorageLevel.MEMORY_ONLY_2)
> // This will pass since the second task attempt will succeed
> assert(rdd.count() === 10)
> // This will fail due to `myAcc.add(100)` won't be executed during the
> second task attempt's
> // execution. Because the second task attempt will load the rdd cache
> directly instead of
> // executing the task function so `myAcc.add(100)` is skipped.
> assert(myAcc.value === 100)
> } {code}
>
> We could also hit this issue with decommission even if the rdd only has one
> copy. For example, decommission could migrate the rdd cache block to another
> executor (the result is actually the same with 2 copies) and the
> decommissioned executor lost before the task reports its success status to
> the driver.
>
> And the issue is a bit more complicated than expected to fix. I have tried to
> give some fixes but all of them are not ideal:
> Option 1: Clean up any rdd cache related to the failed task: in practice,
> this option can already fix the issue in most cases. However, theoretically,
> rdd cache could be reported to the driver right after the driver cleans up
> the failed task's caches due to asynchronous communication. So this option
> can’t resolve the issue thoroughly;
> Option 2: Disallow rdd cache reuse across the task attempts for the same
> task: this option can 100% fix the issue. The problem is this way can also
> affect the case where rdd cache can be reused across the attempts (e.g., when
> there is no accumulator operation in the task), which can have perf
> regression;
> Option 3: Introduce accumulator cache: first, this requires a new framework
> for supporting accumulator cache; second, the driver should improve its logic
> to distinguish whether the accumulator cache value should be reported to the
> user to avoid overcounting. For example, in the case of task retry, the value
> should be reported. However, in the case of rdd cache reuse, the value
> shouldn’t be reported (should it?);
> Option 4: Do task success validation when a task trying to load the rdd
> cache: this way defines a rdd cache is only valid/accessible if the task has
> succeeded. This way could be either overkill or a bit complex (because
> currently Spark would clean up the task state once it’s finished. So we need
> to maintain a structure to know if task once succeeded or not. )
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
-
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-41497) Accumulator undercounting in the case of retry task with rdd cache
[
https://issues.apache.org/jira/browse/SPARK-41497?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17646971#comment-17646971
]
wuyi commented on SPARK-41497:
--
I'm thinking if we could improve the improved Option 4 by changing the rdd
cache reuse condition a bit:
if there're no accumulators (external only probably) values changed after the
rdd computation, then the rdd's cache should be marked as usable/visible no
matter whether the task succeeds or fail.
(let me think further and see if it's doable..)
> Accumulator undercounting in the case of retry task with rdd cache
> --
>
> Key: SPARK-41497
> URL: https://issues.apache.org/jira/browse/SPARK-41497
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
>Affects Versions: 2.4.8, 3.0.3, 3.1.3, 3.2.2, 3.3.1
>Reporter: wuyi
>Priority: Major
>
> Accumulator could be undercounted when the retried task has rdd cache. See
> the example below and you could also find the completed and reproducible
> example at
> [https://github.com/apache/spark/compare/master...Ngone51:spark:fix-acc]
>
> {code:scala}
> test("SPARK-XXX") {
> // Set up a cluster with 2 executors
> val conf = new SparkConf()
> .setMaster("local-cluster[2, 1,
> 1024]").setAppName("TaskSchedulerImplSuite")
> sc = new SparkContext(conf)
> // Set up a custom task scheduler. The scheduler will fail the first task
> attempt of the job
> // submitted below. In particular, the failed first attempt task would
> success on computation
> // (accumulator accounting, result caching) but only fail to report its
> success status due
> // to the concurrent executor lost. The second task attempt would success.
> taskScheduler = setupSchedulerWithCustomStatusUpdate(sc)
> val myAcc = sc.longAccumulator("myAcc")
> // Initiate a rdd with only one partition so there's only one task and
> specify the storage level
> // with MEMORY_ONLY_2 so that the rdd result will be cached on both two
> executors.
> val rdd = sc.parallelize(0 until 10, 1).mapPartitions { iter =>
> myAcc.add(100)
> iter.map(x => x + 1)
> }.persist(StorageLevel.MEMORY_ONLY_2)
> // This will pass since the second task attempt will succeed
> assert(rdd.count() === 10)
> // This will fail due to `myAcc.add(100)` won't be executed during the
> second task attempt's
> // execution. Because the second task attempt will load the rdd cache
> directly instead of
> // executing the task function so `myAcc.add(100)` is skipped.
> assert(myAcc.value === 100)
> } {code}
>
> We could also hit this issue with decommission even if the rdd only has one
> copy. For example, decommission could migrate the rdd cache block to another
> executor (the result is actually the same with 2 copies) and the
> decommissioned executor lost before the task reports its success status to
> the driver.
>
> And the issue is a bit more complicated than expected to fix. I have tried to
> give some fixes but all of them are not ideal:
> Option 1: Clean up any rdd cache related to the failed task: in practice,
> this option can already fix the issue in most cases. However, theoretically,
> rdd cache could be reported to the driver right after the driver cleans up
> the failed task's caches due to asynchronous communication. So this option
> can’t resolve the issue thoroughly;
> Option 2: Disallow rdd cache reuse across the task attempts for the same
> task: this option can 100% fix the issue. The problem is this way can also
> affect the case where rdd cache can be reused across the attempts (e.g., when
> there is no accumulator operation in the task), which can have perf
> regression;
> Option 3: Introduce accumulator cache: first, this requires a new framework
> for supporting accumulator cache; second, the driver should improve its logic
> to distinguish whether the accumulator cache value should be reported to the
> user to avoid overcounting. For example, in the case of task retry, the value
> should be reported. However, in the case of rdd cache reuse, the value
> shouldn’t be reported (should it?);
> Option 4: Do task success validation when a task trying to load the rdd
> cache: this way defines a rdd cache is only valid/accessible if the task has
> succeeded. This way could be either overkill or a bit complex (because
> currently Spark would clean up the task state once it’s finished. So we need
> to maintain a structure to know if task once succeeded or not. )
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
-
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-41497) Accumulator undercounting in the case of retry task with rdd cache
[
https://issues.apache.org/jira/browse/SPARK-41497?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17646969#comment-17646969
]
wuyi commented on SPARK-41497:
--
> do we have a way to do that ?
[~mridulm80] Currently, we only have the mapping between the task and
accumulators. Accumulators are registered to the task via TaskContext.get()
when they deserialize at the executor.
If we could have a way to know which RDD scope the accumulator within when
deserializing, we could set up the mapping between the RDD and accumulators
then. This probably
be the most difficult part.
> Accumulator undercounting in the case of retry task with rdd cache
> --
>
> Key: SPARK-41497
> URL: https://issues.apache.org/jira/browse/SPARK-41497
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
>Affects Versions: 2.4.8, 3.0.3, 3.1.3, 3.2.2, 3.3.1
>Reporter: wuyi
>Priority: Major
>
> Accumulator could be undercounted when the retried task has rdd cache. See
> the example below and you could also find the completed and reproducible
> example at
> [https://github.com/apache/spark/compare/master...Ngone51:spark:fix-acc]
>
> {code:scala}
> test("SPARK-XXX") {
> // Set up a cluster with 2 executors
> val conf = new SparkConf()
> .setMaster("local-cluster[2, 1,
> 1024]").setAppName("TaskSchedulerImplSuite")
> sc = new SparkContext(conf)
> // Set up a custom task scheduler. The scheduler will fail the first task
> attempt of the job
> // submitted below. In particular, the failed first attempt task would
> success on computation
> // (accumulator accounting, result caching) but only fail to report its
> success status due
> // to the concurrent executor lost. The second task attempt would success.
> taskScheduler = setupSchedulerWithCustomStatusUpdate(sc)
> val myAcc = sc.longAccumulator("myAcc")
> // Initiate a rdd with only one partition so there's only one task and
> specify the storage level
> // with MEMORY_ONLY_2 so that the rdd result will be cached on both two
> executors.
> val rdd = sc.parallelize(0 until 10, 1).mapPartitions { iter =>
> myAcc.add(100)
> iter.map(x => x + 1)
> }.persist(StorageLevel.MEMORY_ONLY_2)
> // This will pass since the second task attempt will succeed
> assert(rdd.count() === 10)
> // This will fail due to `myAcc.add(100)` won't be executed during the
> second task attempt's
> // execution. Because the second task attempt will load the rdd cache
> directly instead of
> // executing the task function so `myAcc.add(100)` is skipped.
> assert(myAcc.value === 100)
> } {code}
>
> We could also hit this issue with decommission even if the rdd only has one
> copy. For example, decommission could migrate the rdd cache block to another
> executor (the result is actually the same with 2 copies) and the
> decommissioned executor lost before the task reports its success status to
> the driver.
>
> And the issue is a bit more complicated than expected to fix. I have tried to
> give some fixes but all of them are not ideal:
> Option 1: Clean up any rdd cache related to the failed task: in practice,
> this option can already fix the issue in most cases. However, theoretically,
> rdd cache could be reported to the driver right after the driver cleans up
> the failed task's caches due to asynchronous communication. So this option
> can’t resolve the issue thoroughly;
> Option 2: Disallow rdd cache reuse across the task attempts for the same
> task: this option can 100% fix the issue. The problem is this way can also
> affect the case where rdd cache can be reused across the attempts (e.g., when
> there is no accumulator operation in the task), which can have perf
> regression;
> Option 3: Introduce accumulator cache: first, this requires a new framework
> for supporting accumulator cache; second, the driver should improve its logic
> to distinguish whether the accumulator cache value should be reported to the
> user to avoid overcounting. For example, in the case of task retry, the value
> should be reported. However, in the case of rdd cache reuse, the value
> shouldn’t be reported (should it?);
> Option 4: Do task success validation when a task trying to load the rdd
> cache: this way defines a rdd cache is only valid/accessible if the task has
> succeeded. This way could be either overkill or a bit complex (because
> currently Spark would clean up the task state once it’s finished. So we need
> to maintain a structure to know if task once succeeded or not. )
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
-
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mai
[jira] [Commented] (SPARK-41497) Accumulator undercounting in the case of retry task with rdd cache
[
https://issues.apache.org/jira/browse/SPARK-41497?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17646962#comment-17646962
]
Mridul Muralidharan commented on SPARK-41497:
-
Agree, if we can determine that - do we have a way to do that ?
> Accumulator undercounting in the case of retry task with rdd cache
> --
>
> Key: SPARK-41497
> URL: https://issues.apache.org/jira/browse/SPARK-41497
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
>Affects Versions: 2.4.8, 3.0.3, 3.1.3, 3.2.2, 3.3.1
>Reporter: wuyi
>Priority: Major
>
> Accumulator could be undercounted when the retried task has rdd cache. See
> the example below and you could also find the completed and reproducible
> example at
> [https://github.com/apache/spark/compare/master...Ngone51:spark:fix-acc]
>
> {code:scala}
> test("SPARK-XXX") {
> // Set up a cluster with 2 executors
> val conf = new SparkConf()
> .setMaster("local-cluster[2, 1,
> 1024]").setAppName("TaskSchedulerImplSuite")
> sc = new SparkContext(conf)
> // Set up a custom task scheduler. The scheduler will fail the first task
> attempt of the job
> // submitted below. In particular, the failed first attempt task would
> success on computation
> // (accumulator accounting, result caching) but only fail to report its
> success status due
> // to the concurrent executor lost. The second task attempt would success.
> taskScheduler = setupSchedulerWithCustomStatusUpdate(sc)
> val myAcc = sc.longAccumulator("myAcc")
> // Initiate a rdd with only one partition so there's only one task and
> specify the storage level
> // with MEMORY_ONLY_2 so that the rdd result will be cached on both two
> executors.
> val rdd = sc.parallelize(0 until 10, 1).mapPartitions { iter =>
> myAcc.add(100)
> iter.map(x => x + 1)
> }.persist(StorageLevel.MEMORY_ONLY_2)
> // This will pass since the second task attempt will succeed
> assert(rdd.count() === 10)
> // This will fail due to `myAcc.add(100)` won't be executed during the
> second task attempt's
> // execution. Because the second task attempt will load the rdd cache
> directly instead of
> // executing the task function so `myAcc.add(100)` is skipped.
> assert(myAcc.value === 100)
> } {code}
>
> We could also hit this issue with decommission even if the rdd only has one
> copy. For example, decommission could migrate the rdd cache block to another
> executor (the result is actually the same with 2 copies) and the
> decommissioned executor lost before the task reports its success status to
> the driver.
>
> And the issue is a bit more complicated than expected to fix. I have tried to
> give some fixes but all of them are not ideal:
> Option 1: Clean up any rdd cache related to the failed task: in practice,
> this option can already fix the issue in most cases. However, theoretically,
> rdd cache could be reported to the driver right after the driver cleans up
> the failed task's caches due to asynchronous communication. So this option
> can’t resolve the issue thoroughly;
> Option 2: Disallow rdd cache reuse across the task attempts for the same
> task: this option can 100% fix the issue. The problem is this way can also
> affect the case where rdd cache can be reused across the attempts (e.g., when
> there is no accumulator operation in the task), which can have perf
> regression;
> Option 3: Introduce accumulator cache: first, this requires a new framework
> for supporting accumulator cache; second, the driver should improve its logic
> to distinguish whether the accumulator cache value should be reported to the
> user to avoid overcounting. For example, in the case of task retry, the value
> should be reported. However, in the case of rdd cache reuse, the value
> shouldn’t be reported (should it?);
> Option 4: Do task success validation when a task trying to load the rdd
> cache: this way defines a rdd cache is only valid/accessible if the task has
> succeeded. This way could be either overkill or a bit complex (because
> currently Spark would clean up the task state once it’s finished. So we need
> to maintain a structure to know if task once succeeded or not. )
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
-
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-41497) Accumulator undercounting in the case of retry task with rdd cache
[
https://issues.apache.org/jira/browse/SPARK-41497?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17646919#comment-17646919
]
wuyi commented on SPARK-41497:
--
[~mridulm80] For b) and c), shouldn't we allow T2 to use the result of T1's
cache if rdd1's computation doesn't include any accumulators?
> Accumulator undercounting in the case of retry task with rdd cache
> --
>
> Key: SPARK-41497
> URL: https://issues.apache.org/jira/browse/SPARK-41497
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
>Affects Versions: 2.4.8, 3.0.3, 3.1.3, 3.2.2, 3.3.1
>Reporter: wuyi
>Priority: Major
>
> Accumulator could be undercounted when the retried task has rdd cache. See
> the example below and you could also find the completed and reproducible
> example at
> [https://github.com/apache/spark/compare/master...Ngone51:spark:fix-acc]
>
> {code:scala}
> test("SPARK-XXX") {
> // Set up a cluster with 2 executors
> val conf = new SparkConf()
> .setMaster("local-cluster[2, 1,
> 1024]").setAppName("TaskSchedulerImplSuite")
> sc = new SparkContext(conf)
> // Set up a custom task scheduler. The scheduler will fail the first task
> attempt of the job
> // submitted below. In particular, the failed first attempt task would
> success on computation
> // (accumulator accounting, result caching) but only fail to report its
> success status due
> // to the concurrent executor lost. The second task attempt would success.
> taskScheduler = setupSchedulerWithCustomStatusUpdate(sc)
> val myAcc = sc.longAccumulator("myAcc")
> // Initiate a rdd with only one partition so there's only one task and
> specify the storage level
> // with MEMORY_ONLY_2 so that the rdd result will be cached on both two
> executors.
> val rdd = sc.parallelize(0 until 10, 1).mapPartitions { iter =>
> myAcc.add(100)
> iter.map(x => x + 1)
> }.persist(StorageLevel.MEMORY_ONLY_2)
> // This will pass since the second task attempt will succeed
> assert(rdd.count() === 10)
> // This will fail due to `myAcc.add(100)` won't be executed during the
> second task attempt's
> // execution. Because the second task attempt will load the rdd cache
> directly instead of
> // executing the task function so `myAcc.add(100)` is skipped.
> assert(myAcc.value === 100)
> } {code}
>
> We could also hit this issue with decommission even if the rdd only has one
> copy. For example, decommission could migrate the rdd cache block to another
> executor (the result is actually the same with 2 copies) and the
> decommissioned executor lost before the task reports its success status to
> the driver.
>
> And the issue is a bit more complicated than expected to fix. I have tried to
> give some fixes but all of them are not ideal:
> Option 1: Clean up any rdd cache related to the failed task: in practice,
> this option can already fix the issue in most cases. However, theoretically,
> rdd cache could be reported to the driver right after the driver cleans up
> the failed task's caches due to asynchronous communication. So this option
> can’t resolve the issue thoroughly;
> Option 2: Disallow rdd cache reuse across the task attempts for the same
> task: this option can 100% fix the issue. The problem is this way can also
> affect the case where rdd cache can be reused across the attempts (e.g., when
> there is no accumulator operation in the task), which can have perf
> regression;
> Option 3: Introduce accumulator cache: first, this requires a new framework
> for supporting accumulator cache; second, the driver should improve its logic
> to distinguish whether the accumulator cache value should be reported to the
> user to avoid overcounting. For example, in the case of task retry, the value
> should be reported. However, in the case of rdd cache reuse, the value
> shouldn’t be reported (should it?);
> Option 4: Do task success validation when a task trying to load the rdd
> cache: this way defines a rdd cache is only valid/accessible if the task has
> succeeded. This way could be either overkill or a bit complex (because
> currently Spark would clean up the task state once it’s finished. So we need
> to maintain a structure to know if task once succeeded or not. )
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
-
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-41497) Accumulator undercounting in the case of retry task with rdd cache
[
https://issues.apache.org/jira/browse/SPARK-41497?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17646825#comment-17646825
]
Mridul Muralidharan commented on SPARK-41497:
-
> For example, a task is constructed by `rdd1.cache().rdd2`. So if the task
> fails due to rdd2's computation, I think rdd1's cache should still be able to
> reuse.
We have three cases here.
For some initial task T1:
a) Computation of rdd2 within T1 should have no issues, since initial
computation would result in caching it locally - and rdd2 computation is using
the result of that iterator [1].
b) Failure of T1 and now T2 executes for the same partition - as proposed, T2
will not use result of T1's cache, since it not marked as usable (even though
the block might be still marked cached [2]).
c) Replication and/or decom and T2 runs - same case as (b).
[1] We should special case and allow reads from the same task which cached the
block - even if it has not yet been marked usable.
[2] When T1 failed, we would drop the blocks which got cached due to it. But
even in the case of a race, the flag prevents the use of the cached block.
> Accumulator undercounting in the case of retry task with rdd cache
> --
>
> Key: SPARK-41497
> URL: https://issues.apache.org/jira/browse/SPARK-41497
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
>Affects Versions: 2.4.8, 3.0.3, 3.1.3, 3.2.2, 3.3.1
>Reporter: wuyi
>Priority: Major
>
> Accumulator could be undercounted when the retried task has rdd cache. See
> the example below and you could also find the completed and reproducible
> example at
> [https://github.com/apache/spark/compare/master...Ngone51:spark:fix-acc]
>
> {code:scala}
> test("SPARK-XXX") {
> // Set up a cluster with 2 executors
> val conf = new SparkConf()
> .setMaster("local-cluster[2, 1,
> 1024]").setAppName("TaskSchedulerImplSuite")
> sc = new SparkContext(conf)
> // Set up a custom task scheduler. The scheduler will fail the first task
> attempt of the job
> // submitted below. In particular, the failed first attempt task would
> success on computation
> // (accumulator accounting, result caching) but only fail to report its
> success status due
> // to the concurrent executor lost. The second task attempt would success.
> taskScheduler = setupSchedulerWithCustomStatusUpdate(sc)
> val myAcc = sc.longAccumulator("myAcc")
> // Initiate a rdd with only one partition so there's only one task and
> specify the storage level
> // with MEMORY_ONLY_2 so that the rdd result will be cached on both two
> executors.
> val rdd = sc.parallelize(0 until 10, 1).mapPartitions { iter =>
> myAcc.add(100)
> iter.map(x => x + 1)
> }.persist(StorageLevel.MEMORY_ONLY_2)
> // This will pass since the second task attempt will succeed
> assert(rdd.count() === 10)
> // This will fail due to `myAcc.add(100)` won't be executed during the
> second task attempt's
> // execution. Because the second task attempt will load the rdd cache
> directly instead of
> // executing the task function so `myAcc.add(100)` is skipped.
> assert(myAcc.value === 100)
> } {code}
>
> We could also hit this issue with decommission even if the rdd only has one
> copy. For example, decommission could migrate the rdd cache block to another
> executor (the result is actually the same with 2 copies) and the
> decommissioned executor lost before the task reports its success status to
> the driver.
>
> And the issue is a bit more complicated than expected to fix. I have tried to
> give some fixes but all of them are not ideal:
> Option 1: Clean up any rdd cache related to the failed task: in practice,
> this option can already fix the issue in most cases. However, theoretically,
> rdd cache could be reported to the driver right after the driver cleans up
> the failed task's caches due to asynchronous communication. So this option
> can’t resolve the issue thoroughly;
> Option 2: Disallow rdd cache reuse across the task attempts for the same
> task: this option can 100% fix the issue. The problem is this way can also
> affect the case where rdd cache can be reused across the attempts (e.g., when
> there is no accumulator operation in the task), which can have perf
> regression;
> Option 3: Introduce accumulator cache: first, this requires a new framework
> for supporting accumulator cache; second, the driver should improve its logic
> to distinguish whether the accumulator cache value should be reported to the
> user to avoid overcounting. For example, in the case of task retry, the value
> should be reported. However, in the case of rdd cache reuse, the value
> shouldn’t be reported (should it?);
> Option 4: Do task success validation when
[jira] [Commented] (SPARK-41497) Accumulator undercounting in the case of retry task with rdd cache
[
https://issues.apache.org/jira/browse/SPARK-41497?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17646658#comment-17646658
]
huangtengfei commented on SPARK-41497:
--
I also think that option3/4(include the improved proposal [~mridulm80]
suggested) would be promising, the issue could be resolved either from the
accumulator side or rdd cache side.
And option4 seems more straightforward since it's a complement to existing
cache mechanism. And making decision based on task status could be a feasible
solution. As mentioned above, the downside is that it may be overkill. If such
cases are small probability events, maybe it is also acceptable.
> Accumulator undercounting in the case of retry task with rdd cache
> --
>
> Key: SPARK-41497
> URL: https://issues.apache.org/jira/browse/SPARK-41497
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
>Affects Versions: 2.4.8, 3.0.3, 3.1.3, 3.2.2, 3.3.1
>Reporter: wuyi
>Priority: Major
>
> Accumulator could be undercounted when the retried task has rdd cache. See
> the example below and you could also find the completed and reproducible
> example at
> [https://github.com/apache/spark/compare/master...Ngone51:spark:fix-acc]
>
> {code:scala}
> test("SPARK-XXX") {
> // Set up a cluster with 2 executors
> val conf = new SparkConf()
> .setMaster("local-cluster[2, 1,
> 1024]").setAppName("TaskSchedulerImplSuite")
> sc = new SparkContext(conf)
> // Set up a custom task scheduler. The scheduler will fail the first task
> attempt of the job
> // submitted below. In particular, the failed first attempt task would
> success on computation
> // (accumulator accounting, result caching) but only fail to report its
> success status due
> // to the concurrent executor lost. The second task attempt would success.
> taskScheduler = setupSchedulerWithCustomStatusUpdate(sc)
> val myAcc = sc.longAccumulator("myAcc")
> // Initiate a rdd with only one partition so there's only one task and
> specify the storage level
> // with MEMORY_ONLY_2 so that the rdd result will be cached on both two
> executors.
> val rdd = sc.parallelize(0 until 10, 1).mapPartitions { iter =>
> myAcc.add(100)
> iter.map(x => x + 1)
> }.persist(StorageLevel.MEMORY_ONLY_2)
> // This will pass since the second task attempt will succeed
> assert(rdd.count() === 10)
> // This will fail due to `myAcc.add(100)` won't be executed during the
> second task attempt's
> // execution. Because the second task attempt will load the rdd cache
> directly instead of
> // executing the task function so `myAcc.add(100)` is skipped.
> assert(myAcc.value === 100)
> } {code}
>
> We could also hit this issue with decommission even if the rdd only has one
> copy. For example, decommission could migrate the rdd cache block to another
> executor (the result is actually the same with 2 copies) and the
> decommissioned executor lost before the task reports its success status to
> the driver.
>
> And the issue is a bit more complicated than expected to fix. I have tried to
> give some fixes but all of them are not ideal:
> Option 1: Clean up any rdd cache related to the failed task: in practice,
> this option can already fix the issue in most cases. However, theoretically,
> rdd cache could be reported to the driver right after the driver cleans up
> the failed task's caches due to asynchronous communication. So this option
> can’t resolve the issue thoroughly;
> Option 2: Disallow rdd cache reuse across the task attempts for the same
> task: this option can 100% fix the issue. The problem is this way can also
> affect the case where rdd cache can be reused across the attempts (e.g., when
> there is no accumulator operation in the task), which can have perf
> regression;
> Option 3: Introduce accumulator cache: first, this requires a new framework
> for supporting accumulator cache; second, the driver should improve its logic
> to distinguish whether the accumulator cache value should be reported to the
> user to avoid overcounting. For example, in the case of task retry, the value
> should be reported. However, in the case of rdd cache reuse, the value
> shouldn’t be reported (should it?);
> Option 4: Do task success validation when a task trying to load the rdd
> cache: this way defines a rdd cache is only valid/accessible if the task has
> succeeded. This way could be either overkill or a bit complex (because
> currently Spark would clean up the task state once it’s finished. So we need
> to maintain a structure to know if task once succeeded or not. )
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
-
To unsubscribe, e
[jira] [Commented] (SPARK-41497) Accumulator undercounting in the case of retry task with rdd cache
[
https://issues.apache.org/jira/browse/SPARK-41497?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17646639#comment-17646639
]
wuyi commented on SPARK-41497:
--
[~mridulm80] Sounds like a better idea than option 4. But I think it still
doesn't work well for the case like:
For example, a task is constructed by `rdd1.cache().rdd2`. So if the task fails
due to rdd2's computation, I think rdd1's cache should still be able to reuse.
> Accumulator undercounting in the case of retry task with rdd cache
> --
>
> Key: SPARK-41497
> URL: https://issues.apache.org/jira/browse/SPARK-41497
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
>Affects Versions: 2.4.8, 3.0.3, 3.1.3, 3.2.2, 3.3.1
>Reporter: wuyi
>Priority: Major
>
> Accumulator could be undercounted when the retried task has rdd cache. See
> the example below and you could also find the completed and reproducible
> example at
> [https://github.com/apache/spark/compare/master...Ngone51:spark:fix-acc]
>
> {code:scala}
> test("SPARK-XXX") {
> // Set up a cluster with 2 executors
> val conf = new SparkConf()
> .setMaster("local-cluster[2, 1,
> 1024]").setAppName("TaskSchedulerImplSuite")
> sc = new SparkContext(conf)
> // Set up a custom task scheduler. The scheduler will fail the first task
> attempt of the job
> // submitted below. In particular, the failed first attempt task would
> success on computation
> // (accumulator accounting, result caching) but only fail to report its
> success status due
> // to the concurrent executor lost. The second task attempt would success.
> taskScheduler = setupSchedulerWithCustomStatusUpdate(sc)
> val myAcc = sc.longAccumulator("myAcc")
> // Initiate a rdd with only one partition so there's only one task and
> specify the storage level
> // with MEMORY_ONLY_2 so that the rdd result will be cached on both two
> executors.
> val rdd = sc.parallelize(0 until 10, 1).mapPartitions { iter =>
> myAcc.add(100)
> iter.map(x => x + 1)
> }.persist(StorageLevel.MEMORY_ONLY_2)
> // This will pass since the second task attempt will succeed
> assert(rdd.count() === 10)
> // This will fail due to `myAcc.add(100)` won't be executed during the
> second task attempt's
> // execution. Because the second task attempt will load the rdd cache
> directly instead of
> // executing the task function so `myAcc.add(100)` is skipped.
> assert(myAcc.value === 100)
> } {code}
>
> We could also hit this issue with decommission even if the rdd only has one
> copy. For example, decommission could migrate the rdd cache block to another
> executor (the result is actually the same with 2 copies) and the
> decommissioned executor lost before the task reports its success status to
> the driver.
>
> And the issue is a bit more complicated than expected to fix. I have tried to
> give some fixes but all of them are not ideal:
> Option 1: Clean up any rdd cache related to the failed task: in practice,
> this option can already fix the issue in most cases. However, theoretically,
> rdd cache could be reported to the driver right after the driver cleans up
> the failed task's caches due to asynchronous communication. So this option
> can’t resolve the issue thoroughly;
> Option 2: Disallow rdd cache reuse across the task attempts for the same
> task: this option can 100% fix the issue. The problem is this way can also
> affect the case where rdd cache can be reused across the attempts (e.g., when
> there is no accumulator operation in the task), which can have perf
> regression;
> Option 3: Introduce accumulator cache: first, this requires a new framework
> for supporting accumulator cache; second, the driver should improve its logic
> to distinguish whether the accumulator cache value should be reported to the
> user to avoid overcounting. For example, in the case of task retry, the value
> should be reported. However, in the case of rdd cache reuse, the value
> shouldn’t be reported (should it?);
> Option 4: Do task success validation when a task trying to load the rdd
> cache: this way defines a rdd cache is only valid/accessible if the task has
> succeeded. This way could be either overkill or a bit complex (because
> currently Spark would clean up the task state once it’s finished. So we need
> to maintain a structure to know if task once succeeded or not. )
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
-
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-41497) Accumulator undercounting in the case of retry task with rdd cache
[
https://issues.apache.org/jira/browse/SPARK-41497?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17646477#comment-17646477
]
Mridul Muralidharan commented on SPARK-41497:
-
Agree - there appears to be a bunch of scenarios where this can be triggered.
Essentially, whenever block save succeeds and task itself fails, we can end up
with this scenario.
As detailed, this could be storage level with replication > 1, block
decomissioning, etc - where executor or driver has replicated the data.
I think this can also happen when task itself fails - but after persisting the
data (even without replication or decomm) - for example, due to some
local/transient shuffle write issues (for example).
For the options listed:
Agree, option 1 does not solve the issue.
I am also not inclined towards option 2 due to the potential perf impact -
though I would expect this to be a rare scenario.
Option 3 looks like a very involved approach, and I am not sure if we can cover
all the corner cases.
I am wondering if we can modify option 4 such that it helps.
There are multiple approaches perhaps - one strawman proposal:
a) Add a bit to BlockStatus indicating whether block can be used or not. And
currently this bit gets flipped when the task which computed it successfully
completes.
b) Maintain a taskId -> BlockStatus* mapping - which is cleaned up whenever a
task completes (if successful, then flip bit - else remove its blocks - and
replicas (if any)).
c) Propagate taskId in reportBlockStatus from doPutIterator, etc - where new
block is getting created in the system.
Thoughts ?
> Accumulator undercounting in the case of retry task with rdd cache
> --
>
> Key: SPARK-41497
> URL: https://issues.apache.org/jira/browse/SPARK-41497
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
>Affects Versions: 2.4.8, 3.0.3, 3.1.3, 3.2.2, 3.3.1
>Reporter: wuyi
>Priority: Major
>
> Accumulator could be undercounted when the retried task has rdd cache. See
> the example below and you could also find the completed and reproducible
> example at
> [https://github.com/apache/spark/compare/master...Ngone51:spark:fix-acc]
>
> {code:scala}
> test("SPARK-XXX") {
> // Set up a cluster with 2 executors
> val conf = new SparkConf()
> .setMaster("local-cluster[2, 1,
> 1024]").setAppName("TaskSchedulerImplSuite")
> sc = new SparkContext(conf)
> // Set up a custom task scheduler. The scheduler will fail the first task
> attempt of the job
> // submitted below. In particular, the failed first attempt task would
> success on computation
> // (accumulator accounting, result caching) but only fail to report its
> success status due
> // to the concurrent executor lost. The second task attempt would success.
> taskScheduler = setupSchedulerWithCustomStatusUpdate(sc)
> val myAcc = sc.longAccumulator("myAcc")
> // Initiate a rdd with only one partition so there's only one task and
> specify the storage level
> // with MEMORY_ONLY_2 so that the rdd result will be cached on both two
> executors.
> val rdd = sc.parallelize(0 until 10, 1).mapPartitions { iter =>
> myAcc.add(100)
> iter.map(x => x + 1)
> }.persist(StorageLevel.MEMORY_ONLY_2)
> // This will pass since the second task attempt will succeed
> assert(rdd.count() === 10)
> // This will fail due to `myAcc.add(100)` won't be executed during the
> second task attempt's
> // execution. Because the second task attempt will load the rdd cache
> directly instead of
> // executing the task function so `myAcc.add(100)` is skipped.
> assert(myAcc.value === 100)
> } {code}
>
> We could also hit this issue with decommission even if the rdd only has one
> copy. For example, decommission could migrate the rdd cache block to another
> executor (the result is actually the same with 2 copies) and the
> decommissioned executor lost before the task reports its success status to
> the driver.
>
> And the issue is a bit more complicated than expected to fix. I have tried to
> give some fixes but all of them are not ideal:
> Option 1: Clean up any rdd cache related to the failed task: in practice,
> this option can already fix the issue in most cases. However, theoretically,
> rdd cache could be reported to the driver right after the driver cleans up
> the failed task's caches due to asynchronous communication. So this option
> can’t resolve the issue thoroughly;
> Option 2: Disallow rdd cache reuse across the task attempts for the same
> task: this option can 100% fix the issue. The problem is this way can also
> affect the case where rdd cache can be reused across the attempts (e.g., when
> there is no accumulator operation in the task), which can have perf
> regression;
> Option 3: Int
[jira] [Commented] (SPARK-41497) Accumulator undercounting in the case of retry task with rdd cache
[
https://issues.apache.org/jira/browse/SPARK-41497?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17646097#comment-17646097
]
wuyi commented on SPARK-41497:
--
[~mridulm80] [~tgraves] [~attilapiros] [~ivoson] any good ideas?
> Accumulator undercounting in the case of retry task with rdd cache
> --
>
> Key: SPARK-41497
> URL: https://issues.apache.org/jira/browse/SPARK-41497
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
>Affects Versions: 2.4.8, 3.0.3, 3.1.3, 3.2.2, 3.3.1
>Reporter: wuyi
>Priority: Major
>
> Accumulator could be undercounted when the retried task has rdd cache. See
> the example below and you could also find the completed and reproducible
> example at
> [https://github.com/apache/spark/compare/master...Ngone51:spark:fix-acc]
>
> {code:scala}
> test("SPARK-XXX") {
> // Set up a cluster with 2 executors
> val conf = new SparkConf()
> .setMaster("local-cluster[2, 1,
> 1024]").setAppName("TaskSchedulerImplSuite")
> sc = new SparkContext(conf)
> // Set up a custom task scheduler. The scheduler will fail the first task
> attempt of the job
> // submitted below. In particular, the failed first attempt task would
> success on computation
> // (accumulator accounting, result caching) but only fail to report its
> success status due
> // to the concurrent executor lost. The second task attempt would success.
> taskScheduler = setupSchedulerWithCustomStatusUpdate(sc)
> val myAcc = sc.longAccumulator("myAcc")
> // Initiate a rdd with only one partition so there's only one task and
> specify the storage level
> // with MEMORY_ONLY_2 so that the rdd result will be cached on both two
> executors.
> val rdd = sc.parallelize(0 until 10, 1).mapPartitions { iter =>
> myAcc.add(100)
> iter.map(x => x + 1)
> }.persist(StorageLevel.MEMORY_ONLY_2)
> // This will pass since the second task attempt will succeed
> assert(rdd.count() === 10)
> // This will fail due to `myAcc.add(100)` won't be executed during the
> second task attempt's
> // execution. Because the second task attempt will load the rdd cache
> directly instead of
> // executing the task function so `myAcc.add(100)` is skipped.
> assert(myAcc.value === 100)
> } {code}
>
> We could also hit this issue with decommission even if the rdd only has one
> copy. For example, decommission could migrate the rdd cache block to another
> executor (the result is actually the same with 2 copies) and the
> decommissioned executor lost before the task reports its success status to
> the driver.
>
> And the issue is a bit more complicated than expected to fix. I have tried to
> give some fixes but all of them are not ideal:
> Option 1: Clean up any rdd cache related to the failed task: in practice,
> this option can already fix the issue in most cases. However, theoretically,
> rdd cache could be reported to the driver right after the driver cleans up
> the failed task's caches due to asynchronous communication. So this option
> can’t resolve the issue thoroughly;
> Option 2: Disallow rdd cache reuse across the task attempts for the same
> task: this option can 100% fix the issue. The problem is this way can also
> affect the case where rdd cache can be reused across the attempts (e.g., when
> there is no accumulator operation in the task), which can have perf
> regression;
> Option 3: Introduce accumulator cache: first, this requires a new framework
> for supporting accumulator cache; second, the driver should improve its logic
> to distinguish whether the accumulator cache value should be reported to the
> user to avoid overcounting. For example, in the case of task retry, the value
> should be reported. However, in the case of rdd cache reuse, the value
> shouldn’t be reported (should it?);
> Option 4: Do task success validation when a task trying to load the rdd
> cache: this way defines a rdd cache is only valid/accessible if the task has
> succeeded. This way could be either overkill or a bit complex (because
> currently Spark would clean up the task state once it’s finished. So we need
> to maintain a structure to know if task was successful or not. )
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
-
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
