Intel

Intel ® Software Network

Language:
English

Intel Software TV

Intel Academic Community

Home › Software TV

Preparing Students for Ubiquitous Parallelism Pt. 5 of 6

  • Segment 4 of 6
  • August 24, 2011
  • Paul Steinberg (Intel)
  • Download

Michael Wrinn With the migration to multi-core processors well underway, the academic and commercial communities are becoming alert to the need for a corresponding shift in the approach to software: parallelism is essentially everywhere, in the har

Expand Description Collapse Description

Michael Wrinn

With the migration to multi-core processors well underway, the academic and commercial communities are becoming alert to the need for a corresponding shift in the approach to software: parallelism is essentially everywhere, in the hardware; how ought parallelism influence the teachings in universities? Curricular timeframes make this particularly challenging: given the latencies in developing new material, and the subsequent time before those enlightened students graduate and enter the work force, we are trying to anticipate the state of parallel systems years into the future. Current trends point toward a much larger number of cores, with a mix of specialized processing units, and non-uniform memory access - indeed, such platforms exist today. What is the best foundation for those in computational professions to work well with, say, heterogeneous manycore systems? I am part of a team at Intel chartered to respond to that question. Sitting at the intersection of internal development, university research, and university education, we create and share (for free) training material based on connections found among those domains; faculty may use this content directly in their courseware. We continue to explore, with the academic community, ways to address the broad challenge of "thinking parallel".

 

Michael Wrinn
Intel Corporation
Hillsboro, OR 97124
Michael.Wrinn@intel.com

Leave a Comment

To obtain technical support, please go to Software Support

Date Added | Popularity
Date Series Author
August 24, 2011 Intel Atom Processors in Academia
  • 1
    This class will outline the next generation of Intel Atom processors available from Intel's Embedded and Communications Group. These low power processors are finding increasing use in academic programs, not only as part of the curriculum, but al
  • 2
    This class will outline the next generation of Intel Atom processors available from Intel's Embedded and Communications Group. These low power processors are finding increasing use in academic programs, not only as part of the curriculum, but al
 Jerry Makare (Intel)
August 24, 2011 Gaming Architecture: Why Parallelism is Important
  • 1
    Task and domain decompositions are two basic and powerful approaches of exploiting parallelism in a given problem. An effective decomposition requires careful partitioning of tasks and data, and a balanced mapping of these partitions across the avai
  • 2
    Task and domain decompositions are two basic and powerful approaches of exploiting parallelism in a given problem. An effective decomposition requires careful partitioning of tasks and data, and a balanced mapping of these partitions across the avai
  • 3
    Task and domain decompositions are two basic and powerful approaches of exploiting parallelism in a given problem. An effective decomposition requires careful partitioning of tasks and data, and a balanced mapping of these partitions across the avai
  • 4
    ask and domain decompositions are two basic and powerful approaches of exploiting parallelism in a given problem. An effective decomposition requires careful partitioning of tasks and data, and a balanced mapping of these partitions across the avail
  • 5
    Task and domain decompositions are two basic and powerful approaches of exploiting parallelism in a given problem. An effective decomposition requires careful partitioning of tasks and data, and a balanced mapping of these partitions across the avai
  • 6
    Task and domain decompositions are two basic and powerful approaches of exploiting parallelism in a given problem. An effective decomposition requires careful partitioning of tasks and data, and a balanced mapping of these partitions across the avai
  • 7
    Task and domain decompositions are two basic and powerful approaches of exploiting parallelism in a given problem. An effective decomposition requires careful partitioning of tasks and data, and a balanced mapping of these partitions across the avai
 Jerry Makare (Intel)
August 24, 2011 Suddenly All Computing is Parallel: Seizing Opportunity Amid the Clamor
  • 1
    Michael Wrinn\'s Keynote from SIGCSE 2010. The shift in computing hardware to parallel systems is well underway. Sequential chips are no longer designed, and the proud era of von Neumann architecture passes into history. Foundational change of t
  • 2
    hael Wrinn's Keynote form SIGCSE 2010. The shift in computing hardware to parallel systems is well underway. Sequential chips are no longer designed, and the proud era of von Neumann architecture passes into history. Foundational change of this
  • 3
    Michael Wrinn's keynote from SIGCSE 2010. The shift in computing hardware to parallel systems is well underway. Sequential chips are no longer designed, and the proud era of von Neumann architecture passes into history. Foundational change of th
  • 4
    Michael Wrinn's keynote from SIGCSE 2010. The shift in computing hardware to parallel systems is well underway. Sequential chips are no longer designed, and the proud era of von Neumann architecture passes into history. Foundational change of th
 Jerry Makare (Intel)
1-3 of 3

© Intel Corporation