Method and apparatus for transmitting data elements between threads of a parallel computer system

A computer system comprising:

• a plurality of processors, each processor including a plurality of processor cores;
• a plurality of memories, each memory being associated with a respective processor, each memory including a physical queue that is associated with a processor core of the respective processor and that includes a plurality of sequentially linked memory elements, each memory element storing a data element transmitted between two threads executed on the processor cores and an associated thread index that specifies a source thread including the respective data element, the physical queues together forming a global logical queue of the parallel computer system, the global logical queue including a data element management table storing a count specifying a total number of data elements enqueued by each source thread executed on a processor core and that are located in one of the physical queues of the at least one global logical queue, and a processor core index specifying a specific processor core associated with a physical queue containing the data elements enqueued by a respective source thread.