FAQ:
Java

The Java interface is provided in Open MPI on a provisional basis. It is not part of the current or any proposed MPI standard. Continued inclusion of the Java interface is contingent upon active user interest and continued developer support.

The decision to add a Java interface to Open MPI was motivated by a request from the Hadoop community. Hadoop is a Java-based environment for processing extremely large data sets. Modeled on the Google enterprise system, it has evolved into its own rapidly growing open-source community.

Although executed in parallel, Hadoop processes are independent and thus would only possibly use MPI as a base for efficient messaging. Of greater interest, however, is the ability to add MPI-based follow-on processing to a Hadoop map-reduce operation. Since Hadoop map-reduce is typically done in Java, any subsequent MPI operations would best be done in that language.

The Java interface currently covers all MPI 1.2 APIs. Work to extend the interface to the full MPI-3 specification is on-going.

The Java interface documentation is included in the Open MPI man pages. Please note that we are working to complete this documentation - if you don't see a specific interface documented, it may just be that we haven't gotten to it yet.

The Java interface will build only if it is explicitly requested and the required JDK support is found. Java support requires that Open MPI be built at least with shared libraries (i.e., --enable-shared) - any additional options are fine and will not conflict. Note that this is the default for Open MPI, so you don't have to explicitly add the option.

If the JDK is not in a place where we automatically find it, you can specify the location. Two options are available for this purpose:

--with-jdk-bindir=[the location of javac and javah]
--with-jdk-headers=[the directory containing jni.h]

For simplicity, typical configurations are provided in platform files under contrib/platform/hadoop. These will meet the needs of most users, or at least provide a starting point for your own custom configuration.

In summary, therefore, you can configure the system using the following Java-related options:

./configure --with-platform=contrib/platform/hadoop/[your-platform] ...

or

./configure --enable-mpi-java --with-jdk-bindir=foo --with-jdk-headers=bar ...

or simply

./configure --enable-mpi-java ...

if the JDK is in a standard place that we automatically find.

For convenience, the mpijavac wrapper compiler has been provided for compiling Java-based MPI applications. It ensures that all required MPI libraries and class paths are defined. You can see the actual command line using the --showme option, if you are interested.

Once your application has been compiled, you can run it with the standard mpirun command line:

shell$ mpirun [options] java [your-java-options] my-app-class

For convenience, mpirun has been updated to detect the java command and ensure that the required MPI libraries and class paths are defined to support execution. You therefore do NOT need to specify the Java library path to the MPI installation, nor the MPI classpath. Any class path definitions required for your application should be specified either on the command line or via the CLASSPATH environmental variable. Note that the local directory will be added to the class path if nothing is specified.

As always, the java executable, all required libraries, and your application classes must be available on all nodes.

The Java interface in Open MPI is logically separated from,and completely transparent to, all other Open MPI users and have zero performance impact on the rest of the code/bindings.

We are currently quantifying the performance of MPI-based Java applications. Historically, Java-based applications have exhibited a roughly 20% reduction in performance when compared to the C-based equivalent. However, Java has improved significantly over the years and we expect this difference to be less that previously observed.