Institute of Computer Languages
Compilers and Languages Group

Talks 2007 - Michael Gschwind

The Compilers and Languages Group invites you to a talk given by

Dr. Michael Gschwind

IBM TJ Watson Research Center, NY, USA Web:


Cell Broadband Engine: Architecture and Software

Date: Tuesday, September 4th, 2007
Time: 14:00 (s.t.)
Location: TU Wien, Elektrotechnik, EI 4 Reithoffer--Hörsaal, Gusshausstra├če 25-29 (Altbau), 2. Stock


The Cell Broadband Engine was designed to provide an order of magnitude increase in computing performance over desktop systems. To accomplish this goal in the presence of the diminishing returns of investing in uni-thread performance, the key was to address next generation architecture challenges. Cell B.E. addresses the major pressing issues in modern architecture -- the memory wall, the power wall, and the frequency wall -- by shifting the compute paradigm from a uni-processor design to an efficient chip multiprocessor. We will describe the Cell Broadband Engine Architecture, and how the four dimensions of parallelism, thread level parallelism, instruction level parallelism, data-level parallelism, and data processing/transfer parallelism guided the design of the Cell B.E. architecture to deliver a leap in performance by exploiting application parallelism at all levels. Cell addresses the compute density challenge to increase the performance per area and power, and delivers high performance for compute intensive codes by exploiting chip multiprocessing to address inherent power and memory latency issues.

To increase computational efficiency, the Cell B.E. is based on a heterogeneous chip multiprocessing architecture with a general purpose Power processor element (PPE) and eight attached synergistic processor element (SPE). The SPE is based on a novel pervasively data-parallel SIMD architecture. The Cell B.E. system architecture is optimized to ensure efficient data sharing and synchronization between the PPE and SPEs. We will discuss trade-offs between hardware complexity and software efficiency which lead to our choice of a system-wide shared virtual memory architecture and coherent DMA. We conclude with a discussion of system software and operating system support based on open software, including Cell Linux.

About Dr. Michael Gschwind:

Dr. Michael Gschwind is an IBM Master Inventor, an IBM Power architect and a design lead for a future IBM System. He was one of the initiators and a leading contributor to the Cell Broadband Engine system architecture definition as well as a lead architect of the Synergistic Processor architecture. During the definition of the Synergistic Processor architecture, Michael also developed the first Cell Broadband Engine compiler and served as first technical lead for system software development. Michael joined the IBM TJ Watson Research Center, Yorktown Heights, NY, in 1997. He has held leadership positions in several seminal projects, including the DAISY dynamic compilation project where he was a lead architect for the BOA high-frequency statically scheduled architecture, and was a leading contributor to the development of pioneering dynamic compilation techniques. Michael was also a leading contributor to seminal work on power/performance trade-offs in microprocessor designs which formalized the futility of the frequency-centric uniprocessor design approach used in the industry at the time, an insight that had already guided the design of the Cell Broadband Engine.

Michael's contributions to IBM systems and technology have been recognized with several corporate awards. In addition to his contributions to the design and implementation of IBM systems, he is the author of over 75 papers, covering hardware/software co-design, compiler technology, multimedia processing, and high-performance computer architecture, and has received key patents for his inventions in these areas. In 2006, InformationWeek recognized Michael with an IT Innovator and Influencer Award. In addition to his corporate contributions, Michael has been a faculty member at Technische Universität Wien, Vienna, Austria, and a visiting faculty member at Princeton University where he has taught classes on advanced computer architecture. Michael received PhD and Dipl.-Ing.degrees in computer science from Technische Universit├Ąt Wien, Vienna, Austria. (URL:

   About Us
      Talks 2017
      Talks 2016
      Talks 2015
      Talks 2014
      Talks 2013
      Talks 2012
      Talks 2011
      Talks 2010
      Talks 2009
      Talks 2008
      Talks 2007
      Talks 2006
      Talks 2005
      Talks 2004
Fast Access:
Previous Talk
Next Talk
Faculty of Informatics
Vienna University of Technology
top | HTML 4.01 | last update: 2017-02-13 (Webmaster)