摘要:
A multi-core system and a method for processing data in parallel in the multi-core system are provided. In the multi-core system, partitioning and allocating of data may be dynamically controlled based on local memory information. Thus, it is possible to increase an availability of a Central Processing Unit (CPU) and a local memory, and is possible to improve a performance of data parallel processing.
摘要:
A multi-core system and a method for processing data in parallel in the multi-core system are provided. In the multi-core system, partitioning and allocating of data may be dynamically controlled based on local memory information. Thus, it is possible to increase an availability of a Central Processing Unit (CPU) and a local memory, and is possible to improve a performance of data parallel processing.
摘要:
Provided is a method and apparatus for measuring a progress or a performance of an application program in a computing environment using a micro-architecture. An apparatus for thread progress tracking may select a thread included in an application program, may determine, based on a predetermined criterion, whether an execution scheme for at least one instruction included in the thread corresponds to an effective execution scheme in which an execution time is uniform or a non-effective execution scheme in which a delayed cycle is included and the execution time is non-uniform, and may generate an effective progress index (EPI) by accumulating an execution time of an instruction executed by the effective execution scheme other than an instruction executed by the non-effective execution scheme.
摘要:
A task migration system is provided which transmits a migration request signal for a plurality of first tasks to a migration manager using a resource manager, transmits information used in response to the migration request signal from a migration initiation handler to the migration manager when a first task, of which a migration point is in a capture ready state, among the plurality of first tasks is received from a processor, and captures, using the migration manager, the migration point of the first task in the capture ready state, in response to a migration request signal for the first task in the capture ready state, so that the first task with the captured migration point migrates to a second task.
摘要:
A dynamic task management system and method for data parallel processing on a multi-core system are provided. The dynamic task management system may generate a registration signal for a task to be parallel processed, may generate a dynamic management signal used to dynamically manage at least one task, in response to the generated registration signal, and may control the at least one task to be created or cancelled in at least one core in response to the generated dynamic management signal.
摘要:
A virtual architecture generating apparatus and method, a runtime system, a multi-core system, and methods of operating the runtime system and the multi-core system may include analyzing a requirement of an application, a feature of the application, and a requirement of a system enabling an execution of the application, and include generating a virtual architecture corresponding to the application, based on a physical architecture of a reconfigurable processor, the analyzed requirements and the analyzed feature.
摘要:
Provided is a method and apparatus for measuring a progress or a performance of an application program in a computing environment using a micro-architecture. An apparatus for thread progress tracking may select a thread included in an application program, may determine, based on a predetermined criterion, whether an execution scheme for at least one instruction included in the thread corresponds to an effective execution scheme in which an execution time is uniform or a non-effective execution scheme in which a delayed cycle is included and the execution time is non-uniform, and may generate an effective progress index (EPI) by accumulating an execution time of an instruction executed by the effective execution scheme other than an instruction executed by the non-effective execution scheme.
摘要:
Provided is a method and apparatus for ensuring a deterministic execution characteristic of an application program to perform data processing and execute particular functions in a computing environment using a micro architecture. A lock controlling apparatus based on a deterministic progress index (DPI) may include a loading unit to load a DPI of a first core and a DPI of a second core among DPIs of a plurality of cores at a lock acquisition point in time of each thread, a comparison unit to compare the DPI of the first core and the DPI of the second core, and a controller to assign a lock to a thread of the first core when the DPI of the first core is less than the DPI of the second core and when the second core corresponds to a last core to be compared among the plurality of cores.
摘要:
A multiprocessor system and a migration method of the multiprocessor system are provided. The multiprocessor system may process dynamic data and static data of a task to be operated in another memory or another processor without converting pointers, in a distributed memory environment and in a multiprocessor environment having a local memory, so that dynamic task migration may be realized.
摘要:
Provided is a method and apparatus for measuring a performance or a progress state of an application program to perform data processing and execute particular functions in a computing environment using a micro architecture. A thread progress tracking apparatus may include a selector to select at least one thread constituting an application program; a determination unit to determine, based on a predetermined criterion, whether an instruction execution scheme corresponds to a deterministic execution scheme having a regular cycle or a nondeterministic execution scheme having an irregular delay cycle with respect to each of at least one instruction constituting a corresponding thread; and a deterministic progress counter to generate a deterministic progress index with respect to an instruction that is executed by the deterministic execution scheme, excluding an instruction that is executed by the nondeterministic execution scheme.