java.lang.OutOfMemoryError: Java heap space

[ghost]

Bug Report

Affected tool(s)

Mark Duplicates

Affected version(s)

Picard version: 2.18.11
Using Java 8u151

Description

The error I keep getting is:

INFO 2018-08-29 11:58:12 MarkDuplicates Tracking 2186445 as yet unmatched pairs. 117336 records in RAM.
INFO 2018-08-29 11:58:21 MarkDuplicates Read 489,000,000 records. Elapsed time: 00:58:09s. Time for last 1,000,000: 8s. Last read position: chr8:15,392,138
INFO 2018-08-29 11:58:21 MarkDuplicates Tracking 2186162 as yet unmatched pairs. 114912 records in RAM.
[Wed Aug 29 11:58:39 AEST 2018] picard.sam.markduplicates.MarkDuplicates done. Elapsed time: 58.48 minutes.
Runtime.totalMemory()=954728448
To get help, see http://broadinstitute.github.io/picard/index.html#GettingHelp
Exception in thread "main" java.lang.OutOfMemoryError: Java heap space
at java.lang.reflect.Array.newArray(Native Method)
at java.lang.reflect.Array.newInstance(Array.java:75)
at java.util.Arrays.parallelSort(Arrays.java:1178)
at htsjdk.samtools.util.SortingCollection.spillToDisk(SortingCollection.java:248)
at htsjdk.samtools.util.SortingCollection.add(SortingCollection.java:183)
at picard.sam.markduplicates.MarkDuplicates.buildSortedReadEndLists(MarkDuplicates.java:590)
at picard.sam.markduplicates.MarkDuplicates.doWork(MarkDuplicates.java:232)
at picard.cmdline.CommandLineProgram.instanceMain(CommandLineProgram.java:277)
at picard.cmdline.PicardCommandLine.instanceMain(PicardCommandLine.java:103)

Steps to reproduce

I am using the command:
java -Xmx60g -jar picard.jar MarkDuplicates
INPUT=$BAM_DIR$TUMOR$SORTED
OUTPUT=$BAM_DIR$TUMOR$SUFFIX
METRICS_FILE=$BAM_DIR$TUMOR$SUFFIX.metrics.txt
VALIDATION_STRINGENCY=LENIENT
TMP_DIR=$TMP_DIR

Expected behavior

I ran this exact bam file through 1 month ago with 8gb, same parameters and no problems.

Actual behavior

As above

Thank you for you help!

[yfarjoun]

Try lowering the records in RAM argument

…

On Tue, Aug 28, 2018 at 10:34 PM renny817 ***@***.***> wrote: Bug Report Affected tool(s) Mark Duplicates Affected version(s) Picard version: 2.18.11 Using Java 8u151 Description The error I keep getting is: INFO 2018-08-29 11:58:12 MarkDuplicates Tracking 2186445 as yet unmatched pairs. 117336 records in RAM. INFO 2018-08-29 11:58:21 MarkDuplicates Read 489,000,000 records. Elapsed time: 00:58:09s. Time for last 1,000,000: 8s. Last read position: chr8:15,392,138 INFO 2018-08-29 11:58:21 MarkDuplicates Tracking 2186162 as yet unmatched pairs. 114912 records in RAM. [Wed Aug 29 11:58:39 AEST 2018] picard.sam.markduplicates.MarkDuplicates done. Elapsed time: 58.48 minutes. Runtime.totalMemory()=954728448 To get help, see http://broadinstitute.github.io/picard/index.html#GettingHelp Exception in thread "main" java.lang.OutOfMemoryError: Java heap space at java.lang.reflect.Array.newArray(Native Method) at java.lang.reflect.Array.newInstance(Array.java:75) at java.util.Arrays.parallelSort(Arrays.java:1178) at htsjdk.samtools.util.SortingCollection.spillToDisk(SortingCollection.java:248) at htsjdk.samtools.util.SortingCollection.add(SortingCollection.java:183) at picard.sam.markduplicates.MarkDuplicates.buildSortedReadEndLists(MarkDuplicates.java:590) at picard.sam.markduplicates.MarkDuplicates.doWork(MarkDuplicates.java:232) at picard.cmdline.CommandLineProgram.instanceMain(CommandLineProgram.java:277) at picard.cmdline.PicardCommandLine.instanceMain(PicardCommandLine.java:103) Steps to reproduce I am using the command: java -Xmx60g -jar picard.jar MarkDuplicates INPUT=$BAM_DIR$TUMOR$SORTED OUTPUT=$BAM_DIR$TUMOR$SUFFIX METRICS_FILE=$BAM_DIR$TUMOR$SUFFIX.metrics.txt VALIDATION_STRINGENCY=LENIENT TMP_DIR=$TMP_DIR Expected behavior I ran this exact bam file through 1 month ago with 8gb, same parameters and no problems. Actual behavior As above Thank you for you help! — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#1221>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ACnk0o-IFXA6HwP3uBvwbes4Vcv2QMQzks5uVf2jgaJpZM4WQt_X> .

[ghost]

Done, thank you!

[YONG-JIA]

Hi,
I encountered the same issue, do you mind illustrate how to "Try lowering the records in RAM argument" ? Thanks a in advance.

[wbsimey]

YONG, you can uses the java arg -Xmx2g to increase the amount of memory. I use -Xmx100g on a system with 1TB RAM. Make sure you know how much memory your system has available.
For example:
picard -Xmx100g CollectAlignmentSummaryMetrics ...