Scioto

From HPCRL Wiki

(Difference between revisions)

Latest revision as of 16:25, 7 January 2015

Scioto, or Shared Collections of Task Objects, is a lightweight framework for providing task management on distributed memory machines under one-sided and global-view parallel programming models. Scioto provides locality aware dynamic load balancing and interoperates with MPI, ARMCI, and Global Arrays. Through task parallelism, the Scioto framework provides a solution for overcoming irregularity, load imbalance, and heterogeneity as well as dynamic mapping of computation onto emerging architectures.

Scioto Programming Model

Scioto provides the user with a task pool programming model. Under this model the user expresses their computation as a set of tasks that can be executed in parallel. Tasks read inputs from their task descriptor as well as from data stored in the global address space. Tasks store their output in the global address space and are also permitted to add new tasks to the task collection.

Below is pseudocode that shows the structure of a simple Scioto program.

 // The user decides how to store task arguments within the task
 // body.  Generally, users define a struct to organize this data.
 typedef struct {
   double args;
 } mytask_t;

 // The user provides a function that implements the task.  This
 // function is run by Scioto when executing the task.
 void task_fcn(gtc_t gtc, task_t *my_task) {
   // ...
 }

 int main() {
   gtc_t    tc   = gtc_create(...);      // Create a task collection
   task_t   task = gtc_task_create(...); // Create a task descriptor
   mytask_t my_t = gtc_task_body(task);  // Get the task body
   
   my_t.args = ...;    // Fill in the task's arguments

   gtc_task_add(task); // Add the task to the task collection

   gtc_process(tc);    // Enter task-parallel region
}

Documentation

Programming Interface

Documentation on Scioto's user API can be found here: Scioto API

Publications

Scalable Work Stealing PDF PPT
James Dinan, Sriram Krishnamoorthy, D. Brian Larkins, Jarek Nieplocha, P. Sadayappan
Proc. 21st Intl. Conference on Supercomputing (SC). Portland, OR, Nov. 14-20, 2009.

Scioto: A Framework for Global-View Task Parallelism PDF PPT
James Dinan, Sriram Krishnamoorthy, D. Brian Larkins, Jarek Nieplocha, P. Sadayappan
Proc. of 37th Intl. Conference on Parallel Processing. Portland, OR, Sept. 8-12, 2008.

People

Project Leaders

Contributors

Acknowledgements

This research was supported in part by DOE grant #DE-FC02-06ER25755 and NSF grant #0403342.

Download

Please see the included README file for installation instructions. If you have questions or comments, contact Jim Dinan (dinan at mcs.anl.gov).

Scioto v0.2 -- 11/15/2011
Scioto v0.1 -- 11/15/2010

@@ Line 1: / Line 1: @@
-lRCmj0  <a href="http://aymllqbpihzn.com/">aymllqbpihzn</a>, [url=http://fjimidfbsawu.com/]fjimidfbsawu[/url], [link=http://tjvttoptrllg.com/]tjvttoptrllg[/link], http://yisptxmbddqw.com/
+Scioto, or Shared Collections of Task Objects, is a lightweight framework for providing task management on
+distributed memory machines under one-sided and global-view parallel programming models. Scioto provides locality aware dynamic load balancing and interoperates with MPI, ARMCI, and Global Arrays.  Through task parallelism, the Scioto framework provides a solution for overcoming irregularity, load imbalance, and heterogeneity as well as dynamic mapping of computation onto emerging architectures.
-zKU3R9  <a href="http://tmobnpzbgtyy.com/">tmobnpzbgtyy</a>, [url=http://yugokfujvnpz.com/]yugokfujvnpz[/url], [link=http://pwehspjjyxjl.com/]pwehspjjyxjl[/link], http://pshtnbnkbngy.com/
+== Scioto Programming Model ==
-== Installing Scioto ==
+Scioto provides the user with a task pool programming model.  Under this model the user expresses their computation as a set of ''tasks'' that can be executed in parallel.  Tasks read inputs from their task descriptor as well as from data stored in the global address space.  Tasks store their output in the global address space and are also permitted to add new tasks to the task collection.
-Scioto depends on both [http://www.emsl.pnl.gov/docs/parsoft/armci/ ARMCI] and MPI.  Detailed installation instructions are included in the README file.
+Below is pseudocode that shows the structure of a simple Scioto program.
-=== Installing on a Linux Desktop ===
+<pre>
+ // The user decides how to store task arguments within the task
+ // body.  Generally, users define a struct to organize this data.
+ typedef struct {
+   double args;
+ } mytask_t;
-Scioto is designed for use on distributed memory clusters but it is also possible to try it out on your desktop.  This can also be convenient for development and tends to work best on SMP and multicore systems.  Here's how I do it on my Ubuntu Linux laptop:
+ // The user provides a function that implements the task.  This
+ // function is run by Scioto when executing the task.
+ void task_fcn(gtc_t gtc, task_t *my_task) {
+   // ...
+  }
-# Install MPI.  Under Ubuntu, MPICH for shared memory can be easily installed via:
+  int main() {
-## $ apt-get install mpich-shmem-bin
+   gtc_t    tc   = gtc_create(...);      // Create a task collection
-# Download [http://www.emsl.pnl.gov/docs/parsoft/armci/ ARMCI v1.3.]
+   task_t   task = gtc_task_create(...); // Create a task descriptor
-# Compile ARMCI using your MPI distribution.  Here's how I do it (''note: if you are running a 64 bit OS, you should change the target to LINUX64 and the ARMCI_LIBS path to LINUX64''):
+   mytask_t my_t = gtc_task_body(task);  // Get the task body
-## $ tar xvzf armci-1-3.tgz
-## $ cd armci-1-3
+   my_t.args = ...;    // Fill in the task's arguments
-## $ make CC=mpicc TARGET=LINUX MSG_COMMS=MPI ARMCI_NETWORK=SOCKETS
-# Compile Scioto:
-## $ make CC=mpicc ARMCI_LIBS=~/src/armci-1-3/lib/LINUX/ ARMCI_INCLUDE=~/src/armci-1-3/src/
-# Compile the Scioto examples:
-## $ cd tc-examples
-## $ make CC=mpicc ARMCI_LIBS=~/src/armci-1-3/lib/LINUX/ ARMCI_INCLUDE=~/src/armci-1-3/src/
-# Try it out!  Using Ubuntu's MPICH:
-## $ mpirun.mpich-shmem -n 2 ./test-task
-== Acknowledgements ==
+   gtc_task_add(task); // Add the task to the task collection
-Current project members:
+   gtc_process(tc);    // Enter task-parallel region
-* [http://www.cse.ohio-state.edu/~dinan James Dinan] (Scioto maintainer)
+}
-* [http://www.cse.ohio-state.edu/~larkins D. Brian Larkins]
+</pre>
+== Documentation ==
+=== Programming Interface ===
+Documentation on Scioto's user API can be found here: [[Scioto API]]
+=== Publications ===
+'''Scalable Work Stealing'''
+[http://www.cse.ohio-state.edu/~dinan/research/tpool/dinan_sc09.pdf PDF]
+[http://www.cse.ohio-state.edu/~dinan/research/tpool/dinan_sc09_slides.ppt PPT]
+<br />
+James Dinan, Sriram Krishnamoorthy, D. Brian Larkins, Jarek Nieplocha, P. Sadayappan <br />
+Proc. 21st Intl. Conference on Supercomputing (SC). Portland, OR, Nov. 14-20, 2009.
+'''Scioto: A Framework for Global-View Task Parallelism'''
+[http://www.cse.ohio-state.edu/~dinan/research/tpool/dinan_icpp08.pdf PDF]
+[http://www.cse.ohio-state.edu/~dinan/research/tpool/dinan_icpp08_slides.ppt PPT]
+<br />
+James Dinan, Sriram Krishnamoorthy, D. Brian Larkins, Jarek Nieplocha, P. Sadayappan <br />
+Proc. of 37th Intl. Conference on Parallel Processing. Portland, OR, Sept. 8-12, 2008.
+== People ==
+=== Project Leaders ===
+* [http://www.mcs.anl.gov/~dinan James Dinan] (Primary contact)
 * [http://www.cse.ohio-state.edu/~saday Prof. P. Sadayappan]
 * [http://hpc.pnl.gov/people/sriram/ Sriram Krishnamoorthy]
+=== Contributors ===
+* [http://www.cse.ohio-state.edu/~larkins D. Brian Larkins]
 * [http://hpc.pnl.gov/people/jarek/ Jarek Nieplocha]
-This research was supported in part by DOE grant ''#DE-FC02-06ER25755'' and NSF
+== Acknowledgements ==
-grant ''#0403342''.
+This research was supported in part by DOE grant ''#DE-FC02-06ER25755'' and NSF grant ''#0403342''.
 == Download ==
-We are currently providing early development snapshots for download.  Please contact Jim Dinan (dinan at cse.ohio-state.edu) if you have questions or comments regarding Scioto.
+Please see the included README file for installation instructions.  If you have questions or comments, contact Jim Dinan (dinan at mcs.anl.gov).
-* Dev Snapshot, 09-13-2008 [http://www.cse.ohio-state.edu/~dinan/scioto/scioto-dev-09-13-2008.tar.gz]
+* [http://hpcrl.cse.ohio-state.edu/scioto/scioto-r7688.tar.gz Scioto v0.2] -- ''11/15/2011''
-* Dev Snapshot, 09-10-2008 [http://www.cse.ohio-state.edu/~dinan/scioto/scioto-dev-09-10-2008.tar.gz]
+* [http://www.mcs.anl.gov/~dinan/download/scioto/scioto-0.1.tar.gz Scioto v0.1] -- ''11/15/2010''

Scioto

Latest revision as of 16:25, 7 January 2015

Contents

Scioto Programming Model

Documentation

Programming Interface

Publications

People

Project Leaders

Contributors

Acknowledgements

Download

Views

Personal tools

Navigation

Search

Toolbox