SC is the International Conference for
High Performance Computing, Networking,
Storage and Analysis



SCHEDULE: NOV 12-18, 2011

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.

You can also create your personal schedule on the SC11 app (Boopsie) on your smartphone. Simply select a session you want to attend and "add" it to your plan. Continue in this manner until you have created your own personal schedule. All your events will appear under "My Event Planner" on your smartphone.

Time Iterated Dataflow: A Dataflow Execution Model for HPC Programs

SESSION: Doctoral Research Showcase (1 of 2)

EVENT TYPE: Doctoral Research Showcase

TIME: 2:15PM - 2:30PM

SESSION CHAIR: Volodymyr Kindratenko

Presenter(s):Daniel Orozco

ROOM:TCC LL1

ABSTRACT:
This work presents the Time Iterated Dataflow (TIEFlow) execution model. The TIEFlow model extends the Dynamic Dataflow model to efficiently develop and execute HPC programs in multicore processors. The TIEFlow model addresses the fact that parallel loops are abundant in HPC programs by expresses programs as graphs where each node represents a parallel loop. Program graphs in TIEFlow use weighted arcs to represent loop carried dependencies. The resulting graph model allows simple representation of commonly used constructs such as task parallelism, task pipelining, and program composability. TIEFlow has been implemented and has resulted in a number of research and commercial tools including the SWARM system currently offered by ET International, a fully distributed runtime system that does not require support from an OS and a programming model with its associated compiler. Preliminary results where sample HPC kernels were developed show that TIEFlow is appropriate to develop and execute HPC programs.

Chair/Presenter Details:

Volodymyr Kindratenko (Chair) - NCSA

Daniel Orozco - University of Delaware

Add to iCal  Click here to download .ics calendar file

Add to Outlook  Click here to download .vcs calendar file

Add to Google Calendarss  Click here to add event to your Google Calendar

   Sponsors    ACM    IEEE