Synchronization method

A computer system with multiple cores arranged to implement a synchronization method, the system comprising:

• a group of cores arranged to allow parallel execution of a plurality of threads using barrier synchronization, where each thread in the group waits for all other threads at a barrier before progressing, the cores are arranged to allow execution of the threads until a first thread reaches the barrier;
• a core on which the first thread is executing is arranged to then allow the first thread to enter a polling state, in which the first thread checks for a release condition indicating an end of the barrier;
• and
• further cores are arranged to allow execution of subsequent threads which reach the barrier subsequently, to allow the subsequent threads to be moved to the core on which the first thread is executing and then to be powered down as a number of moved threads increases;
• wherein the core on which the first thread is executing is arranged to allow the first thread to detect the release condition, and to allow the subsequent threads to move back to the further cores, and the further cores are operable to be powered up for use by the threads.

A non-transitory, machine-readable storage medium storing a computer program which when executed on a computer system with multiple cores carries out a method, the method comprising:

• executing a group of threads in parallel on a plurality of cores, the group of threads being synchronized using barrier synchronization in which each thread in the group waits for all other threads at a barrier before progressing, where the group of threads executes until a first thread reaches the barrier;
• repeatedly checking, when the first thread enters a polling state, for a release condition indicating an end of the barrier;
• moving subsequent threads to reach the barrier to a core on which the first thread is executing;
• and
• powering down other cores as a number of moved threads increases, and
• wherein when the first thread detects the release condition, powered down cores are powered up and are available for use by the threads.

A synchronization method in a computer system with multiple cores, comprising:

• executing a group of processes in parallel on a plurality of cores, the group of processes being synchronized using barrier synchronization in which each process in the group waits for all other processes at a barrier before progressing;
• executing the group of processes until a first process reaches the barrier;
• repeatedly checking, when the first process enters a polling state, for a release condition indicating an end of the barrier;
• moving subsequent processes to reach the barrier to a core on which the first process is executing;
• and
• powering down other cores down as a number of moved processes increases and
• wherein when the first process detects the release condition, powered down cores are powered up and are available for use by the processes.

A computer system with multiple cores arranged to implement a synchronization method, comprising:

• a group of cores arranged to allow parallel execution of a plurality of processes using barrier synchronization, where each process in the group waits for all other processes at a barrier before progressing, the cores are arranged to allow execution of the processes until a first process reaches the barrier;
• a core on which the first process is executing is arranged to then allow the first process to enter a polling state, in which the first process checks for a release condition indicating an end of the barrier;
• and
• further cores are arranged to allow execution of subsequent processes which reach the barrier subsequently;
• to allow the subsequent processes to be moved to the core on which the first process is executing and then to be powered down as a number of moved processes increases;
• wherein the core on which the first process is executing is arranged to allow the first process to detect the release condition, and to allow the subsequent processes to move back to the further cores, and the further cores are operable to be powered up for use by the processes.

A non-transitory, machine-readable storage medium storing a computer program which when executed on a computer system with multiple cores carries out a method, comprising:

• executing a group of processes in parallel on a plurality of cores, the group of processes being synchronized using barrier synchronization in which each process in the group waits for all other processes at a barrier before progressing;
• executing the group of processes until a first process reaches the barrier;
• repeatedly checking, when the first process enters a polling state, for a release condition indicating an end of the barrier;
• moving subsequent processes to reach the barrier to a core on which the first process is executing;
• and
• powering other cores as a number of moved processes increases, and
• wherein when the first process detects the release condition, powered down cores are powered up and are available for use by the processes.