[Pl-seminar] Reminder: 10/13 Seminar: Thomas Wahl, Behavioral Non-Portability in Decision-Making Programs

Daniel Patterson dbp at ccs.neu.edu
Thu Oct 6 19:37:58 EDT 2016


Reminder that this is NEXT week!

On Thu, Sep 29, 2016 at 4:55 PM, Daniel Patterson <dbp at ccs.neu.edu> wrote:

> NUPRL Seminar presents
>
> Thomas Wahl
> Northeastern University
>
> 12:00pm-1:30pm
> Thursday, Oct. 13 2016
> Room 366 WVH (http://www.ccs.neu.edu/home/wand/directions.html)
>
>
> Behavioral Non-Portability in Decision-Making Programs
>
> Abstract:
>
> The precise semantics of floating-point arithmetic programs depends on the
> execution platform, including the compiler and the target hardware. Such
> platform dependencies infringe on the highly desirable goal of software
> portability (which is in fact promised by heterogeneous computing
> frameworks like OpenCL): the same program run on the same inputs on
> different platforms can produce different results. In other words,
> portability does not guarantee reproducibility, and this is a more or less
> accepted state of affairs.
>
> Serious doubts on the portability of numeric applications arise when
> differences in results are behavioral, i.e. when they lead to changes in
> the control flow of a program. In this talk I will first present an
> algorithm that takes a numeric procedure and determines whether a given
> input can lead to different decisions depending merely on how the
> arithmetic in the procedure is compiled and executed. I will then show how
> this algorithm can be used in static and dynamic analyses of programs, to
> estimate their numeric stability. I will illustrate the results on examples
> characteristic of numeric computing where control flow divergence actually
> occurs across different execution platforms.
>
> Joint with Yijia Gu, Mahsa Bayati, and Miriam Leeser, Northeastern
> University, Boston, USA
>
> Bio:
> Thomas Wahl joined the faculty of Northeastern University in 2011. His
> research concerns the reliability (whatever that means) of complex
> computing systems. Two domains notorious for their fragility are
> concurrency and numerical computing. With colleagues, Wahl has investigated
> how floating-point arithmetic can "hijack" a program's computation when run
> on non-standard architectures, such as heterogeneous and custom-made
> embedded platforms. You will witness some hijacking attempts in the talk
> today.
>
>
-------------- next part --------------
HTML attachment scrubbed and removed


More information about the pl-seminar mailing list