Scioto

From HPCRL Wiki
(Difference between revisions)
Jump to: navigation, search
Line 1: Line 1:
Scioto (Shared Collections of Task Objects) is a lightweight
+
== Scioto: Shared Collections of Task Objects ==
framework for providing task management on distributed memory machines under
+
 
one-sided and global-view parallel programming models.  Scioto provides
+
Scioto (pronounced /sigh-OH-toe/ [http://en.wikipedia.org/wiki/Scioto_River]) is a framework for supporting task parallelism on distributed memory machines under one-sided and global view parallel programming models.  This framework allows the user to express their program as a set of tasks that are automatically scheduled and load balanced with respect to locality by the runtime system.   Through task parallelism, the Scioto framework provides an approach to overcoming irregularity, load imbalance, and heterogeneity as well as dynamic mapping of computation onto emerging architectures.
locality aware dynamic load balancing and interoperates with MPI, ARMCI, and
+
 
Global Arrays. Additionally, Scioto's task model and programming
+
Scioto uses ARMCI and MPI internally.  We have tested interoperability with MPI, ARMCI, and Global Arrays applications and are investigating interoperability with additional parallel programming tools.
interface are compatible with many other existing parallel models including
+
UPC, SHMEM, and CAF.  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.
+
  
 
More details coming soon...
 
More details coming soon...

Revision as of 21:37, 10 September 2008

Scioto: Shared Collections of Task Objects

Scioto (pronounced /sigh-OH-toe/ [1]) is a framework for supporting task parallelism on distributed memory machines under one-sided and global view parallel programming models. This framework allows the user to express their program as a set of tasks that are automatically scheduled and load balanced with respect to locality by the runtime system. Through task parallelism, the Scioto framework provides an approach to overcoming irregularity, load imbalance, and heterogeneity as well as dynamic mapping of computation onto emerging architectures.

Scioto uses ARMCI and MPI internally. We have tested interoperability with MPI, ARMCI, and Global Arrays applications and are investigating interoperability with additional parallel programming tools.

More details coming soon...

Personal tools