[jira] [Commented] (SPARK-15796) Spark 1.6 default memory settings can cause heavy GC when caching

["Sean Owen (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/SPARK-15796?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15318109#comment-15318109 ] 

Sean Owen commented on SPARK-15796:
-----------------------------------

You're mostly saying that if you cache too much (or allow too much to be cached) you run out of memory. This is of course going to happen, so I don't know what you expect there.

I agree about the old generation size, though you're referring to defaults on one JVM version. It wouldn't be crazy to change this default though if this is a prevalent default.

If you mean you have to increase the amount of memory for caching and increase the old gen size if you intend to use almost all your memory for caching with pretty good GC behavior -- yes. That's normal for the JVM.

storageFraction controls the minimum memory reserved for caching, yes. If you want to cache more (at the expense of memory for execution of tasks) you increase it. Yes it's the reverse of how similar settings used to work. Now you say how much you want to allocate to a need, at minimum, and leave the rest to everything else. As opposed to grabbing some fraction of the whole, which either made you waste some resources (sum = 100%) or over-commit (sum > 100%) and possibly run out of memory.

> Spark 1.6 default memory settings can cause heavy GC when caching
> -----------------------------------------------------------------
>
>                 Key: SPARK-15796
>                 URL: https://issues.apache.org/jira/browse/SPARK-15796
>             Project: Spark
>          Issue Type: Bug
>    Affects Versions: 1.6.0, 1.6.1
>            Reporter: Gabor Feher
>
> While debugging performance issues in a Spark program, I've found a simple way to slow down Spark 1.6 significantly by filling the RDD memory cache. This seems to be a regression, because setting "spark.memory.useLegacyMode=true" fixes the problem. Here is a repro that is just a simple program that fills the memory cache of Spark using a MEMORY_ONLY cached RDD (but of course this comes up in more complex situations, too):
> {code}
> import org.apache.spark.SparkContext
> import org.apache.spark.SparkConf
> import org.apache.spark.storage.StorageLevel
> object CacheDemoApp { 
>   def main(args: Array[String]) {
>     val conf = new SparkConf().setAppName("Cache Demo Application")                                       
>     val sc = new SparkContext(conf)
>     val startTime = System.currentTimeMillis()
>                                                                                                           
>     val cacheFiller = sc.parallelize(1 to 500000000, 1000)                                                
>       .mapPartitionsWithIndex {
>         case (ix, it) =>
>           println(s"CREATE DATA PARTITION ${ix}")                                                         
>           val r = new scala.util.Random(ix)
>           it.map(x => (r.nextLong, r.nextLong))
>       }
>     cacheFiller.persist(StorageLevel.MEMORY_ONLY)
>     cacheFiller.foreach(identity)
>     val finishTime = System.currentTimeMillis()
>     val elapsedTime = (finishTime - startTime) / 1000
>     println(s"TIME= $elapsedTime s")
>   }
> }
> {code}
> If I call it the following way, it completes in around 5 minutes on my Laptop, while often stopping for slow Full GC cycles. I can also see with jvisualvm (Visual GC plugin) that the old generation of JVM is 96.8% filled.
> {code}
> sbt package
> ~/spark-1.6.0/bin/spark-submit \
>   --class "CacheDemoApp" \
>   --master "local[2]" \
>   --driver-memory 3g \
>   --driver-java-options "-XX:+PrintGCDetails" \
>   target/scala-2.10/simple-project_2.10-1.0.jar
> {code}
> If I add any one of the below flags, then the run-time drops to around 40-50 seconds and the difference is coming from the drop in GC times:
>   --conf "spark.memory.fraction=0.6"
> OR
>   --conf "spark.memory.useLegacyMode=true"
> OR
>   --driver-java-options "-XX:NewRatio=3"
> All the other cache types except for DISK_ONLY produce similar symptoms. It looks like that the problem is that the amount of data Spark wants to store long-term ends up being larger than the old generation size in the JVM and this triggers Full GC repeatedly.
> I did some research:
> * In Spark 1.6, spark.memory.fraction is the upper limit on cache size. It defaults to 0.75.
> * In Spark 1.5, spark.storage.memoryFraction is the upper limit in cache size. It defaults to 0.6 and...
> * http://spark.apache.org/docs/1.5.2/configuration.html even says that it shouldn't be bigger than the size of the old generation.
> * On the other hand, OpenJDK's default NewRatio is 2, which means an old generation size of 66%. Hence the default value in Spark 1.6 contradicts this advice.
> http://spark.apache.org/docs/1.6.1/tuning.html recommends that if the old generation is running close to full, then setting spark.memory.storageFraction to a lower value should help. I have tried with spark.memory.storageFraction=0.1, but it still doesn't fix the issue. This is not a surprise: http://spark.apache.org/docs/1.6.1/configuration.html explains that storageFraction is not an upper-limit but a lower limit-like thing on the size of Spark's cache. The real upper limit is spark.memory.fraction.
> To sum up my questions/issues:
> * At least http://spark.apache.org/docs/1.6.1/tuning.html should be fixed. Maybe the old generation size should also be mentioned in configuration.html near spark.memory.fraction.
> * Is it a goal for Spark to support heavy caching with default parameters and without GC breakdown? If so, then better default values are needed.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org