SCHEDULE: NOV 15-20, 2015

When viewing the Technical Program schedule, on the far righthand side is a column labeled "PLANNER." Use this planner to build your own schedule. Once you select an event and want to add it to your personal schedule, just click on the calendar icon of your choice (outlook calendar, ical calendar or google calendar) and that event will be stored there. As you select events in this manner, you will have your own schedule to guide you through the week.

Scalable Sparse Tensor Decompositions in Distributed Memory Systems

SESSION: Tensor Computation

EVENT TYPE: Papers

EVENT TAG(S): Algorithms, Scientific Computing, Analytics

TIME: 4:00PM - 4:30PM

SESSION CHAIR(S): David Abramson

AUTHOR(S):Oguz Kaya, Bora Ucar

ROOM:18CD

ABSTRACT:

We investigate efficient parallelization of the most common iterative sparse tensor decomposition algorithms on distributed memory systems. A key operation in each iteration of these algorithms is the matricized tensor times Khatri-Rao product (MTTKRP), which amounts to element-wise vector multiplication and reduction depending on the sparsity of the tensor. We investigate fine and coarse-grain task definitions for this operation, and propose hypergraph partitioning-based methods for these to achieve load balance as well as reduce the communication requirements. We also design distributed memory sparse tensor library, HyperTensor, which implements a well-known algorithm for the CANDECOMP/PARAFAC(CP) decomposition utilizing these task definitions and partitions. We use this library to test the scalability of the proposed implementation of MTTKRP in CP decomposition context upto 1024 MPI ranks. We observed up to 194 fold speedups using 512 MPI processes on a real world data, and significantly better scalability than a state of the art implementation.

Chair/Author Details:

David Abramson (Chair) - University of Queensland|

Oguz Kaya - ENS Lyon

Bora Ucar - ENS Lyon

  Click here to download .ics calendar file

  Click here to download .vcs calendar file

  Click here to add event to your Google Calendar

Paper provided by the ACM Digital Library

Paper also available from IEEE Computer Society