Fast prediction of shared memory access pattern

A non-transitory machine-readable storage medium that stores a program of executable instructions, where execution of the program results in performing operations that comprise:

• instrumenting events of shared memory accesses in an application program, where the application program is to be executed on a target configuration comprising p nodes;
• executing the application program using p1 processing nodes, where p1 is less than p and satisfies a constraint;
• and
• for accesses made by the executing application program, determining a target thread and mapping determined target threads to either a remote node or a local node corresponding to a remote memory access and to a local memory access, respectively;
• where instrumenting instruments a compiled version of the application program;
• and
• where mapping comprises intercepting shared memory access function calls generated by a compiler and analyzing arguments of the function calls to determine whether an access is a remote access or a local access.

A data processing system comprising at least one processor connected with at least one memory that stores a program of executable instructions, where execution of the program by said at least one processor results in performing operations that comprise:

• instrumenting events of shared memory accesses of a compiled application program written in a programming language compliant with a partitioned global address space paradigm, where the application program is to be executed on a target configuration comprising p nodes and where the events comprise read from shared memory and write to shared memory;
• executing the application program using p1 processing nodes, where p1 is less than p and satisfies a constraint;
• and
• for accesses made by the executing application program, determining a target thread and mapping determined target threads to either a remote node or a local node corresponding to a remote memory access and to a local memory access, respectively;
• where instrumenting instruments a compiled version of the application program, and where execution of the program by said at least one processor when mapping the determined target threads uses a weak symbol compiler mechanism, intercepts shared memory access function calls generated by a compiler, and analyzes arguments of the function calls to determine whether an access is a remote access or a local access.