Sonal Goyal wrote:
Adarsh,
Are you trying to distribute both the native library and the jcuda.jar?
Could you please explain your job's dependencies?
Yes Of course , I am trying to run a Juda program in Hadoop Cluster as I
am able to run it simple through simple javac & java commands at
standalone machine by setting PATH & LD_LIBRARY_PATH varibale to
*/usr/local/cuda/lib* & */home/hadoop/project/jcuda_1.1_linux *folder.
I listed the contents & jars in these directories :
[hadoop@cuda1 lib]$ pwd
/usr/local/cuda/lib
[hadoop@cuda1 lib]$ ls -ls
total 158036
4 lrwxrwxrwx 1 root root 14 Feb 23 19:37 libcublas.so ->
libcublas.so.3
4 lrwxrwxrwx 1 root root 19 Feb 23 19:37 libcublas.so.3 ->
libcublas.so.3.2.16
81848 -rwxrwxrwx 1 root root 83720712 Feb 23 19:37 libcublas.so.3.2.16
4 lrwxrwxrwx 1 root root 14 Feb 23 19:37 libcudart.so ->
libcudart.so.3
4 lrwxrwxrwx 1 root root 19 Feb 23 19:37 libcudart.so.3 ->
libcudart.so.3.2.16
424 -rwxrwxrwx 1 root root 423660 Feb 23 19:37 libcudart.so.3.2.16
4 lrwxrwxrwx 1 root root 13 Feb 23 19:37 libcufft.so ->
libcufft.so.3
4 lrwxrwxrwx 1 root root 18 Feb 23 19:37 libcufft.so.3 ->
libcufft.so.3.2.16
27724 -rwxrwxrwx 1 root root 28351780 Feb 23 19:37 libcufft.so.3.2.16
4 lrwxrwxrwx 1 root root 14 Feb 23 19:37 libcurand.so ->
libcurand.so.3
4 lrwxrwxrwx 1 root root 19 Feb 23 19:37 libcurand.so.3 ->
libcurand.so.3.2.16
4120 -rwxrwxrwx 1 root root 4209384 Feb 23 19:37 libcurand.so.3.2.16
4 lrwxrwxrwx 1 root root 16 Feb 23 19:37 libcusparse.so ->
libcusparse.so.3
4 lrwxrwxrwx 1 root root 21 Feb 23 19:37 libcusparse.so.3 ->
libcusparse.so.3.2.16
43048 -rwxrwxrwx 1 root root 44024836 Feb 23 19:37 libcusparse.so.3.2.16
172 -rwxrwxrwx 1 root root 166379 Nov 25 11:29
libJCublas-linux-x86_64.so
152 -rwxrwxrwx 1 root root 144179 Nov 25 11:29
libJCudaDriver-linux-x86_64.so
16 -rwxrwxrwx 1 root root 8474 Mar 31 2009 libjcudafft.so
136 -rwxrwxrwx 1 root root 128672 Nov 25 11:29
libJCudaRuntime-linux-x86_64.so
80 -rwxrwxrwx 1 root root 70381 Mar 31 2009 libjcuda.so
44 -rwxrwxrwx 1 root root 38039 Nov 25 11:29 libJCudpp-linux-x86_64.so
44 -rwxrwxrwx 1 root root 38383 Nov 25 11:29 libJCufft-linux-x86_64.so
48 -rwxrwxrwx 1 root root 43706 Nov 25 11:29
libJCurand-linux-x86_64.so
140 -rwxrwxrwx 1 root root 133280 Nov 25 11:29
libJCusparse-linux-x86_64.so
And the second folder as :
[hadoop@cuda1 jcuda_1.1_linux64]$ pwd
/home/hadoop/project/hadoop-0.20.2/jcuda_1.1_linux64
[hadoop@cuda1 jcuda_1.1_linux64]$ ls -ls
total 200
8 drwxrwxrwx 6 hadoop hadoop 4096 Feb 24 01:44 doc
8 drwxrwxrwx 3 hadoop hadoop 4096 Feb 24 01:43 examples
32 -rwxrwxr-x 1 hadoop hadoop 28484 Feb 24 01:43 jcuda.jar
4 -rw-rw-r-- 1 hadoop hadoop 0 Mar 1 21:27 libcublas.so.3
4 -rw-rw-r-- 1 hadoop hadoop 0 Mar 1 21:27 libcublas.so.3.2.16
4 -rw-rw-r-- 1 hadoop hadoop 0 Mar 1 21:27 libcudart.so.3
4 -rw-rw-r-- 1 hadoop hadoop 0 Mar 1 21:27 libcudart.so.3.2.16
4 -rw-rw-r-- 1 hadoop hadoop 0 Mar 1 21:27 libcufft.so.3
4 -rw-rw-r-- 1 hadoop hadoop 0 Mar 1 21:27 libcufft.so.3.2.16
4 -rw-rw-r-- 1 hadoop hadoop 0 Mar 1 21:27 libcurand.so.3
4 -rw-rw-r-- 1 hadoop hadoop 0 Mar 1 21:27 libcurand.so.3.2.16
4 -rw-rw-r-- 1 hadoop hadoop 0 Mar 1 21:27 libcusparse.so.3
4 -rw-rw-r-- 1 hadoop hadoop 0 Mar 1 21:27 libcusparse.so.3.2.16
16 -rwxr-xr-x 1 hadoop hadoop 8474 Mar 1 04:12 libjcudafft.so
80 -rwxr-xr-x 1 hadoop hadoop 70381 Mar 1 04:11 libjcuda.so
8 -rwxrwxr-x 1 hadoop hadoop 811 Feb 24 01:43 README.txt
8 drwxrwxrwx 2 hadoop hadoop 4096 Feb 24 01:43 resources
[hadoop@cuda1 jcuda_1.1_linux64]$
I think Hadoop would not able to recognize *jcuda.jar* in Tasktracker
process. Please guide me how to make it available in it.
Thanks & best Regards,
Adrash Sharma
Thanks and Regards,
Sonal
<https://github.com/sonalgoyal/hiho>Hadoop ETL and Data
Integration<https://github.com/sonalgoyal/hiho>
Nube Technologies <http://www.nubetech.co>
<http://in.linkedin.com/in/sonalgoyal>
On Mon, Feb 28, 2011 at 6:54 PM, Adarsh Sharma <[email protected]>wrote:
Sonal Goyal wrote:
Hi Adarsh,
I think your mapred.cache.files property has an extra space at the end.
Try
removing that and let us know how it goes.
Thanks and Regards,
Sonal
<https://github.com/sonalgoyal/hiho>Hadoop ETL and Data
Integration<https://github.com/sonalgoyal/hiho>
Nube Technologies <http://www.nubetech.co>
<http://in.linkedin.com/in/sonalgoyal>
Thanks a Lot Sonal but it doesn't succeed.
Please if possible tell me the proper steps that are need to be followed
after Configuring Hadoop Cluster.
I don't believe that a simple commands succeeded as
[root@cuda1 hadoop-0.20.2]# javac EnumDevices.java
[root@cuda1 hadoop-0.20.2]# java EnumDevices
Total number of devices: 1
Name: Tesla C1060
Version: 1.3
Clock rate: 1296000 MHz
Threads per block: 512
but in Map-reduce job it fails :
11/02/28 18:42:47 INFO mapred.JobClient: Task Id :
attempt_201102281834_0001_m_000001_2, Status : FAILED
java.lang.RuntimeException: java.lang.reflect.InvocationTargetException
at
org.apache.hadoop.util.ReflectionUtils.newInstance(ReflectionUtils.java:115)
at org.apache.hadoop.mapred.MapTask.runNewMapper(MapTask.java:569)
at org.apache.hadoop.mapred.MapTask.run(MapTask.java:305)
at org.apache.hadoop.mapred.Child.main(Child.java:170)
Caused by: java.lang.reflect.InvocationTargetException
at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native
Method)
at
sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:39)
at
sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:27)
at java.lang.reflect.Constructor.newInstance(Constructor.java:513)
at
org.apache.hadoop.util.ReflectionUtils.newInstance(ReflectionUtils.java:113)
... 3 more
Caused by: java.lang.UnsatisfiedLinkError: no jcuda in java.library.path
at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1734)
at java.lang.Runtime.loadLibrary0(Runtime.java:823)
at java.lang.System.loadLibrary(System.java:1028)
at jcuda.driver.CUDADriver.<clinit>(CUDADriver.java:909)
at jcuda.CUDA.init(CUDA.java:62)
at jcuda.CUDA.<init>(CUDA.java:42)
Thanks & best Regards,
Adarsh Sharma
On Mon, Feb 28, 2011 at 5:06 PM, Adarsh Sharma <[email protected]
wrote:
Thanks Sanjay, it seems i found the root cause.
But I result in following error:
[hadoop@ws37-mah-lin hadoop-0.20.2]$ bin/hadoop jar wordcount1.jar
org.myorg.WordCount /user/hadoop/gutenberg /user/hadoop/output1
Exception in specified URI's java.net.URISyntaxException: Illegal
character
in path at index 36: hdfs://192.168.0.131:54310/jcuda.jar
at java.net.URI$Parser.fail(URI.java:2809)
at java.net.URI$Parser.checkChars(URI.java:2982)
at java.net.URI$Parser.parseHierarchical(URI.java:3066)
at java.net.URI$Parser.parse(URI.java:3014)
at java.net.URI.<init>(URI.java:578)
at
org.apache.hadoop.util.StringUtils.stringToURI(StringUtils.java:204)
at
org.apache.hadoop.filecache.DistributedCache.getCacheFiles(DistributedCache.java:593)
at
org.apache.hadoop.mapred.JobClient.configureCommandLineOptions(JobClient.java:638)
at
org.apache.hadoop.mapred.JobClient.submitJobInternal(JobClient.java:761)
at org.apache.hadoop.mapreduce.Job.submit(Job.java:432)
at org.apache.hadoop.mapreduce.Job.waitForCompletion(Job.java:447)
at org.myorg.WordCount.main(WordCount.java:59)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at
sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
at
sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
at java.lang.reflect.Method.invoke(Method.java:597)
at org.apache.hadoop.util.RunJar.main(RunJar.java:156)
Exception in thread "main" java.lang.NullPointerException
at org.apache.hadoop.fs.FileSystem.get(FileSystem.java:176)
at
org.apache.hadoop.filecache.DistributedCache.getTimestamp(DistributedCache.java:506)
at
org.apache.hadoop.mapred.JobClient.configureCommandLineOptions(JobClient.java:640)
at
org.apache.hadoop.mapred.JobClient.submitJobInternal(JobClient.java:761)
at org.apache.hadoop.mapreduce.Job.submit(Job.java:432)
at org.apache.hadoop.mapreduce.Job.waitForCompletion(Job.java:447)
at org.myorg.WordCount.main(WordCount.java:59)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at
sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
at
sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
at java.lang.reflect.Method.invoke(Method.java:597)
at org.apache.hadoop.util.RunJar.main(RunJar.java:156)
Please check my attached mapred-site.xml
Thanks & best regards,
Adarsh Sharma
Kaluskar, Sanjay wrote:
You will probably have to use distcache to distribute your jar to all
the nodes too. Read the distcache documentation; Then on each node you
can add the new jar to the java.library.path through
mapred.child.java.opts.
You need to do something like the following in mapred-site.xml, where
fs-uri is the URI of the file system (something like
host.mycompany.com:54310).
<property>
<name>mapred.cache.files</name>
<value>hdfs://fs-uri/jcuda/jcuda.jar#jcuda.jar </value>
</property>
<property>
<name>mapred.create.symlink</name>
<value>yes</value>
</property>
<property>
<name>mapred.child.java.opts</name>
<value>-Djava.library.path=jcuda.jar</value>
</property>
-----Original Message-----
From: Adarsh Sharma [mailto:[email protected]] Sent: 28 February
2011 16:03
To: [email protected]
Subject: Setting java.library.path for map-reduce job
Dear all,
I want to set some extra jars in java.library.path , used while running
map-reduce program in Hadoop Cluster.
I got a exception entitled "no jcuda in java.library.path" in each map
task.
I run my map-reduce code by below commands :
javac -classpath
/home/hadoop/project/hadoop-0.20.2/hadoop-0.20.2-core.jar://home/hadoop/
project/hadoop-0.20.2/jcuda_1.1_linux64/jcuda.jar:/home/hadoop/project/h
adoop-0.20.2/lib/commons-cli-1.2.jar
-d wordcount_classes1/ WordCount.java
jar -cvf wordcount1.jar -C wordcount_classes1/ .
bin/hadoop jar wordcount1.jar org.myorg.WordCount /user/hadoop/gutenberg
/user/hadoop/output1
Please guide how to achieve this.
Thanks & best Regards,
Adarsh Sharma
