Abstract:
PROBLEM TO BE SOLVED: To provide an optimization process capable of using a unique characteristic of a multiprocessor system for codes of a wide range of categories. SOLUTION: A method (and a system) for transparent dynamic optimization in a multiprocessing environment includes a step of monitoring the execution of an application on a first processor with an execution monitor operating on other processor of the system, and a step of transparently optimizing one or two or more segments of the original application with a runtime optimizer being executed on other processor of the system. COPYRIGHT: (C)2004,JPO
Abstract:
PROBLEM TO BE SOLVED: To provide a method and a structure for effectively emulating the memory matching behavior of a multiprocessing system on another multiprocessing system when a host multiprocessing system supports a mild matching model and a target multiprocessing system designates a strong matching model. SOLUTION: The method (and a system) for emulation in a multiprocessor system includes a step of executing the emulation in which the host mutiprocessing system of the multiprocessor system supports the weak matching model and the target multiprocessing system of the multiprocessor system supports the strong matching model. COPYRIGHT: (C)2004,JPO
Abstract:
PROBLEM TO BE SOLVED: To provide a method and a structure capable of effectively mapping memory address designation of a multiprocessing system for emulating by using virtual memory address designation of another multiprocessing system. SOLUTION: The method (and a system), for emulating the memory address designation of a target system by using the virtual/actual memory mapping mechanism of the operating system of a host mutiprocessor system, includes a step in which a target virtual memory address is inputted into a simulated page table to acquire a host virtual memory address. The target system is not acknowledged by software which the system is operating. COPYRIGHT: (C)2004,JPO
Abstract:
PROBLEM TO BE SOLVED: To provide a method and structure for efficiently emulating the memory consistency behavior of a multiprocessing system on another multiprocessing system when a host multiprocessing system supports a relaxed consistency model and a target multiprocessing system designates a strong consistency model. SOLUTION: A method (and system) of emulation in a multiprocessor system, includes a step of performing an emulation in which the host multiprocessing system of the multiprocessor system supports a weak consistency model, and the target multiprocessing system of the multiprocessor system supports the strong consistency model. COPYRIGHT: (C)2007,JPO&INPIT
Abstract:
PROBLEM TO BE SOLVED: To provide a method and structure which efficiently emulate memory consistency behavior of a certain multiprocessing system on other multiprocessing system when a host multiprocessing system supports a loose consistency model and a target multiprocessing system designates a rigid consistency model. SOLUTION: A method (and system) of emulation in a multiprocessor system includes a step executing an emulation in which a host multiprocessing system of the multiprocessor system supports a loose consistency model, and the target multiprocessing system of the multiprocessor system supports a rigid consistency model. COPYRIGHT: (C)2010,JPO&INPIT
Abstract:
PROBLEM TO BE SOLVED: To provide a method and a structure which permit a mutiprocessing system to use any processor instruction set and a memory architecture in order to efficiently emulate the behavior of another multiprocessing system using any other processor instruction. SOLUTION: The method (and a system), for executing a mutiprocessor program created for a target instruction set architecture on a host computing system having a plurality of processors designed for processing an instruction of second instruction set architecture, includes a step which expresses each part of a program designed for being operated on one processor of a target computing system as one or two or more program threads to be executed on a host computing system. COPYRIGHT: (C)2004,JPO
Abstract:
Verfahren und Systeme zur Optimierung einer Anwendung für ein Computersystem mit mehreren verschiedenen Speicherorten, die durch einen oder mehrere Kommunikationskanäle miteinander verbunden sind, beinhalten ein Bestimmen einer oder mehrerer Datenverarbeitungseigenschaften für einen Datenbereich in einer Anwendung. Eine oder mehrere Datenverarbeitungsrichtlinien für den Datenbereich werden basierend auf den einen oder mehreren Datenverarbeitungseigenschaften festgelegt. Für einen Bereich in der Anwendung, der den Datenbereich an verschiedenen Speicherorten verwendet, werden basierend auf den einen oder mehreren Datenverarbeitungseigenschaften Dateneinrichtungskosten bestimmt. Die Anwendung wird in Übereinstimmung mit den einen oder mehreren Datenverarbeitungsrichtlinien und den Dateneinrichtungskosten für die verschiedenen Speicherorte optimiert.
Abstract:
A computer system includes a condition register having multiple fields. Each field may be used as an independent condition register. A compiler which generates executable code for the computer system assigns instructions to different fields, allowing condition values to be saved while other conditions are evaluated and allowing consecutive instructions which generate or test condition values to be overlapped in their execution.
Abstract:
A computer system includes a condition register having multiple fields. Each field may be used as an independent condition register. A compiler which generates executable code for the computer system assigns instructions to different fields, allowing condition values to be saved while other conditions are evaluated and allowing consecutive instructions which generate or test condition values to be overlapped in their execution.