Search

نمایش تعداد 1-5 از 5

Toward the efficient use of multiple explicitly managed memory subsystems

نوع: Conference Paper

نویسنده : Pena, Antonio J.; Balaji, Pavan

ناشر: IEEE

سال: 2014

Simplifying the Recovery Model of User-Level Failure Mitigation

نوع: Conference Paper

نویسنده : Bland, Wesley; Raffenetti, Kenneth; Balaji, Pavan

ناشر: IEEE

سال: 2014

Processing MPI Derived Datatypes on Noncontiguous GPU-Resident Data

نوع: Journal Paper

نویسنده : Jenkins, J.; Dinan, James; Balaji, Pavan; Peterka, Tom; Samatova, N.F.; Thakur, Rahul

ناشر: IEEE

سال: 2014

[IEEE 2012 IEEE International Conference on Cluster Computing (CLUSTER) - Beijing, China (2012.09.24-2012.09.28)] 2012 IEEE International Conference on Cluster Computing - Enabling Fast, Noncontiguous GPU Data Movement in Hybrid MPI+GPU Environments

نوع: Journal Paper

نویسنده : Jenkins, John; Dinan, James; Balaji, Pavan; Samatova, Nagiza F.; Thakur, Rajeev

سال: 2012

Request PDF

MPI + MPI: a new hybrid approach to parallel programming with MPI plus shared memory

نوع: Journal Paper

نویسنده : Hoefler, Torsten; Dinan, James; Buntinas, Darius; Balaji, Pavan; Barrett, Brian; Brightwell, Ron; Gropp, William; Kale, Vivek; Thakur, Rajeev

سال: 2013

Request PDF

Search

Toward the efficient use of multiple explicitly managed memory subsystems

Simplifying the Recovery Model of User-Level Failure Mitigation

Processing MPI Derived Datatypes on Noncontiguous GPU-Resident Data

[IEEE 2012 IEEE International Conference on Cluster Computing (CLUSTER) - Beijing, China (2012.09.24-2012.09.28)] 2012 IEEE International Conference on Cluster Computing - Enabling Fast, Noncontiguous GPU Data Movement in Hybrid MPI+GPU Environments

MPI + MPI: a new hybrid approach to parallel programming with MPI plus shared memory

نویسنده

ناشر

سال

کلیدواژه

نوع

زبان

نوع محتوا

عنوان ناشر

Search

Filters

Toward the efficient use of multiple explicitly managed memory subsystems

Simplifying the Recovery Model of User-Level Failure Mitigation

Processing MPI Derived Datatypes on Noncontiguous GPU-Resident Data

[IEEE 2012 IEEE International Conference on Cluster Computing (CLUSTER) - Beijing, China (2012.09.24-2012.09.28)] 2012 IEEE International Conference on Cluster Computing - Enabling Fast, Noncontiguous GPU Data Movement in Hybrid MPI+GPU Environments

MPI + MPI: a new hybrid approach to parallel programming with MPI plus shared memory