Global Arrays

Global Arrays was developed in the early 1990s by researchers at the Pacific Northwest National Laboratory. It was created to address the need for efficient and scalable shared-memory programming models on distributed-memory systems, enabling scientific computations to be performed more effectively by allowing direct access to distributed data without relying on traditional message-passing methods.

Strengths of Global Arrays include efficient direct access to distributed data, ease of use for parallel programming, and support for large, multidimensional arrays. Weaknesses involve complexity in debugging and potential scalability issues on extremely large systems. Competitors include MPI (Message Passing Interface), OpenMP, and UPC (Unified Parallel C).

A Global Arrays expert must have proficiency in parallel programming, strong understanding of distributed-memory systems, experience with C/C++ and Fortran, familiarity with MPI for interoperability, and skills in debugging and optimizing high-performance computing applications.