SC13 Home > SC13 Schedule > SC13 Presentation -

SCHEDULE: NOV 16-22, 2013

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.

SESSION: Accelerating Improvements in HPC Application I/O Performance and Efficiency

EVENT TYPE: Emerging Technologies

TIME: 2:50PM - 3:10PM

Presenter(s):Paul Grun

ROOM:Booth 3947

ABSTRACT:
The OpenFabrics Alliance has recently formed a working group with the task of focusing on the evolution of I/O for exascale computing, networking and storage. The newly formed OpenFrameWork Working Group will be applying the principles of Application-centric I/O as it sets about the task of refining the interfaces an application uses to access communications services. Application-centric I/O is defined as the art and science of defining an I/O system that maximizes application effectiveness. Since an I/O system consists of both a set of application interfaces and an underlying communications service, the working group will be examining both areas beginning with a close examination of characteristic application communications requirements. Clearly, achieving the objective maximizing application efficiency and scalability - will require significant collaboration between the open source I/O software developers and the community of application developers and system users. This collaboration is essential since the design of efficient application interfaces is dependent on being responsive to communications requirements at the application level. In this talk we will briefly discuss some motivations that are driving the next level of evolution in application communications and the basic architectural concepts being deployed by the new OFWG. This talk will be of particular interest to developers working in shared memory architectures such as PGAS or SHMEM as well as classical MPI and to system architects focusing on future I/O needs for exascale.

Chair/Presenter Details:

Paul Grun - Cray Inc.

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