View source for MPI+Argobots

As core number of many-core processors keeps increasing, MPI+X is becoming a promising programming model for large scale SMP clusters. It has the potential to utilizing both intra-node and inter-node parallelism with appropriate execution unit and granularity.

[https://collab.mcs.anl.gov/display/ARGOBOTS/Argobots+Home Argobots] is a low-level threading/task infrastructure developed by a joint effort of Argonne National Laboratory, University of Illinois at Urbana-Champaign, University of Tennessee, Knoxville and Pacific Northwest National Laboratory. It provides a lightweight execution model that combines low-latency thread and task scheduling with optimized data-movement functionality.

A benefit of Argobots is providing asynchrony/overlap to MPI. The idea is to make multiple MPI blocking calls at the same time in multiple ULTs, if one MPI call is blocked in ULT A, MPI runtime will detect it and context switch to another ULT to make progress on other blocking calls. Once other ULTs finished their execution, they will switch back to ULT A to continue its execution. In this way, we can keep the CPU busy doing useful work instead of waiting the blocking call.

However, the two-level parallelism of MPI+X introduces new problems such as lock contention in MPI between threads. To avoid unnecessary locks between execution units, MPI+Argobots will explicitly control the context switch between User Level Threads (ULT) and Execution Streams (ES). When switching between ULTs in the same ES, no lock is needed.

== New Thread Level for ULT: MPI_THREAD_ULT ==
We propose another thread level for MPI and thread integration: MPI_THREAD_ULT. In this level, there will be only one ES per process and multiple ULTs in the ES. Because ULTs do not execute concurrently, so there is no lock needed when enter or exit MPI calls. On the other side, when yielding, the current ULT will yield to other ULTs in the same ES, compared to yielding to other ESs with MPI_THREAD_MULTIPLE.

<pre>
MPI_Init_thread(&argc, &argv, MPI_THREAD_ULT, &provided);
</pre>

== Build MPI+Argobots ==

Git repos:
{| class="wikitable"
|-
| Argobot read-only clone URL
| git://git.mcs.anl.gov/argo/argobots.git
|-
| mpich-dev read-only clone URL
| git://git.mpich.org/mpich-dev.git
|}

=== Build Argobots ===
Follow the instructions in https://collab.mcs.anl.gov/display/ARGOBOTS/Getting+and+Building to build Argobots.

<pre>
$ export INSTALL_PATH=/path/to/install
$ git clone --origin argobots git://git.mcs.anl.gov/argo/argobots.git argobots
$ cd argobots
$ ./autogen.sh
$ ./configure --prefix=$INSTALL_PATH
$ make -j 4
$ make install
</pre>

=== Build MPICH ===
MPI+Argobots is currently under develop in mpich-dev repository. To get the source code, do
<pre>
$ git clone --origin mpich-dev git://git.mpich.org/mpich-dev.git mpich-dev
$ cd mpich-dev
$ git checkout mpi-argobots
</pre>

Set paths to link Argobots library.
<pre>
export LD_LIBRARY_PATH=$INSTALL_PATH/lib:$LD_LIBRARY_PATH
export LIBRARY_PATH=$INSTALL_PATH/lib:$LIBRARY_PATH
export C_INCLUDE_PATH=$INSTALL_PATH/include:$C_INCLUDE_PATH
</pre>

Compile.
<pre>
$ ./autogen.sh
$ CFLAGS="-I$INSTALL_PATH/include" ./configure --prefix=$INSTALL_PATH --enable-threads=multiple --with-thread-package=argobots
$ make -j 8
$ make install
</pre>

=== Build and Run MPI+Argobots Examples ===
Set path to use the newly install mpicc and mpiexec.
<pre>
export PATH=$INSTALL_PATH/bin:$PATH
</pre>

Run examples.
<pre>
cd mpich-dev/test/mpi/threads/argobots
make
mpiexec -n 2 ./hello_abt
</pre>

== Example of MPI+Argobots ==
<pre>
MPI_Init_thread();
/* Argobots calls */
MPI_Finalize();
</pre>

ABT_init is included in MPI_Init_thread and ABT_finalize is included in MPI_Finalize. So be careful not to use Argobots calls after MPI_Finalize().