摘要:
The present invention provides a method and apparatus for QR-factorizing matrix on a multiprocessor system, wherein the multiprocessor system comprises at least one core processor and a plurality of accelerators, the method comprises the steps of: iteratively factorizing each panel in the matrix until the whole matrix is factorized; wherein in each iteration, the method comprises: partitioning an unprocessed matrix part in the matrix into a plurality of blocks according to a predetermined block size; partitioning a current processed panel in the unprocessed matrix part into at least two sub panels, wherein the current processed panel is composed of a plurality of blocks; and performing QR factorization one by one on the at least two sub panels with the plurality of accelerators, and updating the data of the sub panel(s) on which no QR factorization has been performed among the at least two sub panels by using the factorization result. The present invention enables a multiprocessor system having a high computing capability to be applied to the matrix QR factorization having a large amount of computation tasks.
摘要:
A method and apparatus for QR-factorizing matrix on a multiprocessor system, wherein the multiprocessor system comprises at least one core processor and a plurality of accelerators, comprises the steps of: iteratively factorizing each panel in the matrix until the whole matrix is factorized; wherein in each iteration, the method comprises: partitioning an unprocessed matrix part in the matrix into a plurality of blocks according to a predetermined block size; partitioning a current processed panel in the unprocessed matrix part into at least two sub panels, wherein the current processed panel is composed of a plurality of blocks; and performing QR factorization one by one on the at least two sub panels with the plurality of accelerators, and updating the data of the sub panel(s) on which no QR factorization has been performed among the at least two sub panels by using the factorization result.
摘要:
The present invention provides a method and apparatus for QR-factorizing matrix on a multiprocessor system, wherein the multiprocessor system comprises at least one core processor and a plurality of accelerators, the method comprises the steps of: iteratively factorizing each panel in the matrix until the whole matrix is factorized; wherein in each iteration, the method comprises: partitioning an unprocessed matrix part in the matrix into a plurality of blocks according to a predetermined block size; partitioning a current processed panel in the unprocessed matrix part into at least two sub panels, wherein the current processed panel is composed of a plurality of blocks; and performing QR factorization one by one on the at least two sub panels with the plurality of accelerators, and updating the data of the sub panel(s) on which no QR factorization has been performed among the at least two sub panels by using the factorization result. The present invention enables a multiprocessor system having a high computing capability to be applied to the matrix QR factorization having a large amount of computation tasks.
摘要:
A mechanism is provided for invoking multi-library application on a multiple processor system, wherein the multiple processor system comprises a Power Processing Element (PPE) and a plurality of Synergistic Processing Element (SPE). Applications including multi-libraries run in the memory of the PPEs. The mechanism comprises maintaining the status of each SPE in the applications running on the PPE, where there are SPE agents for capturing the instructions from the PPE in the SPEs that have been started. In response to a request for invoking a library, the PPE determines whether the number of available SPEs for invoking the library is adequate based on the current status of SPEs. If the number of available SPEs is adequate, the PPE sends a run instruction to selected SPEs. After finishing the invocation of all libraries, the PPE sends termination instructions to all started SPEs.
摘要:
A mechanism is provided for invoking a multi-library application on a multiple processor system, wherein the multiple processor system comprises a Power Processing Element (PPE) and a plurality of Synergistic Processing Element (SPE). Applications including multi-libraries run in the memory of the PPE. The mechanism comprises maintaining the status of each SPE in the application running on the PPE, where there are SPE agents for capturing the instructions from the PPE in the SPEs that have been started. In response to a request for invoking a library, the PPE determines whether the number of available SPEs for invoking the library is adequate based on the current status of SPEs. If the number of available SPEs is adequate, the PPE sends a run instruction to selected SPEs. After finishing the invocation of all libraries, the PPE sends termination instructions to all started SPEs. IBM confidential
摘要:
A mechanism is provided for invoking multi-library application on a multiple processor system, wherein the multiple processor system comprises a Power Processing Element (PPE) and a plurality of Synergistic Processing Element (SPE). Applications including multi-libraries run in the memory of the PPEs. The mechanism comprises maintaining the status of each SPE in the applications running on the PPE, where there are SPE agents for capturing the instructions from the PPE in the SPEs that have been started. In response to a request for invoking a library, the PPE determines whether the number of available SPEs for invoking the library is adequate based on the current status of SPEs. If the number of available SPEs is adequate, the PPE sends a run instruction to selected SPEs. After finishing the invocation of all libraries, the PPE sends termination instructions to all started SPEs.
摘要:
A method and apparatus for QR-factorizing matrix on a multiprocessor system, wherein the multiprocessor system comprises at least one core processor and a plurality of accelerators, comprises the steps of: iteratively factorizing each panel in the matrix until the whole matrix is factorized; wherein in each iteration, the method comprises: partitioning an unprocessed matrix part in the matrix into a plurality of blocks according to a predetermined block size; partitioning a current processed panel in the unprocessed matrix part into at least two sub panels, wherein the current processed panel is composed of a plurality of blocks; and performing QR factorization one by one on the at least two sub panels with the plurality of accelerators, and updating the data of the sub panel(s) on which no QR factorization has been performed among the at least two sub panels by using the factorization result.
摘要:
A method and device for integrating multiple threat security services are disclosed. The method may comprise parsing an incoming packet at a current layer and analyzing the packet with respect to multiple threat security services and so that one or more threat security services needed by the packet may be determined. According to an exemplary embodiment, the current layer may be a layer in a protocol stack constructed based on the multiple threat security services. With this method, integrated multiple threat security services may filter application data and parse network packet data via a single integrated entity, and thus the efficacy of filtering application data may be improved while computation overhead may be reduced.
摘要:
The present invention provides techniques for performing molecular dynamics simulation on a multiprocessor system. The method comprises: dividing a substance space on which molecular dynamics simulation are to be performed into a plurality of cells; storing data of molecules of the plurality of cells in the main memory of the multiprocessor system such that data of molecules of each cell are continuously stored in a memory area corresponding to the cell; and the plurality of accelerators repeatedly acquiring the data of molecules of the plurality of cells from the main memory and performing molecular dynamics simulation computations in parallel such that data of molecules of at least one cell are acquired in one DMA operation. By continuously storing data of molecules of each cell in a memory area corresponding to the cell, the present invention reduces the data exchanges between each accelerator and the main memory during simulation.
摘要:
A method and device for integrating multiple threat security services are disclosed. The method may comprise parsing an incoming packet at a current layer and analyzing the packet with respect to multiple threat security services and so that one or more threat security services needed by the packet may be determined. According to an exemplary embodiment, the current layer may be a layer in a protocol stack constructed based on the multiple threat security services. With this method, integrated multiple threat security services may filter application data and parse network packet data via a single integrated entity, and thus the efficacy of filtering application data may be improved while computation overhead may be reduced.