ZPL uses the array abstraction to implement a
data parallel programming model. This is the reason why ZPL achieves such good performance: having no
parallel directives or other forms of explicit parallelism, ZPL exploits the operational trait that when aggregate computations are described in terms of arrays, many scalar operations must be (implicitly) performed to implement the array operations. This
implied computation can be automatically allotted to different processors to achieve concurrency: Parallelism arises from the semantics of the array operations. ZPL is translated into a conventional
abstract syntax tree representation on which
program analysis and
program optimizations are performed.
ANSI C code is generated as the object code. This C program (which is machine independent because it implements certain operations in abstract form) is then compiled using the native C compiler on the
target machine with custom libraries optimized to the specific platform. The creators of ZPL were: Brad Chamberlain, Sung-Eun Choi, E Christopher Lewis, Calvin Lin, Jason Secosky, Larry Snyder, and W. Derrick Weathersby with assistance from Ruth Anderson, A.J. Bernheim, Marios Dikaiakos, George Forman, and Kurt Partridge. ==See also==