> >CRPC-TR93298-S: Automatic Data Layout for Distributed-Memory Machines
> >                in the D Programming Environment



> >Keywords:
> >
> >  interactive programming environment; Fortran D; automatic data layout;
> >  automatic alignment; dynamic remapping; automatic distribution; 
> >  distributed-memory machines; search space approach; 
> >
> >
> >Abstract:
> >
> >Although distributed-memory message-passing parallel computers are
> >among the most cost-effective high performance machines available,
> >scientists find them extremely difficult to program.
> >Most programmers feel uncomfortable working with a distributed-memory
> >programming model that requires explicit management of local name spaces.
> >To address this problem, researchers have proposed using
> >languages based on a global name space annotated with directives
> >specifying how the data should be mapped onto a distributed memory machine.
> >Using these annotations, a sophisticated compiler can automatically
> >transform a code into a message-passing program suitable for
> >execution on a distributed-memory machine.
> >The Fortran77D and Fortran90D languages support this programming style.
> >Given a Fortran D program, the compiler uses data layout directives to 
> >automatically generate a single-program, multiple data (SPMD) node program 
> >for a given distributed-memory target machine.
> >
> >To achieve high performance with such programs, programmers must select
> >a good data layout.  Current tools provide little or no support for this 
> >selection process. This paper describes an automatic data layout strategy 
> >being investigated for use in the D programming tools currently under
> >development at Rice University. The proposed technique considers the 
> >profitability of dynamic data remapping as it explores a rich search space 
> >of reasonable alignment and distribution schemes.
> >
> >
> >Appeared:
> >     
> >     Proceedings of the First International Workshop on Automatic Distributed
> >     Memory Parallelization, Automatic Data Distribution and Automatic
> >     Parallel Performance Prediction (AP'93)
> >