摘要:
A hierarchical server system efficiently balances the processing load thereon and for shortening the processing time therein is provided, such as a web server system. A system controller, a load balancing device, and a shared memory are provided in a multi-layer server system made of a plurality of servers. When processing implemented with a first layer server is temporarily stopped in order to acquire information needed for processing from a second layer server, process information needed to resume the processing is recorded in the shared memory. When the necessary information is sent back to the first layer server, the system controller inquires about work statuses of all first layer servers to select another first layer server to resume the processing based upon the inquiry results. The then selected first layer server then resumes the processing using the information that was sent back and the process information in the shared memory.
摘要:
A main logical unit and a standby logical unit are defined by a process controller in a shared main memory multiprocessor, and an information storage space accessible from both logical units is provided. The main logical unit stores address information onto that information storage space by indicating a memory area it controls as the main memory area. When failover or cloning becomes necessary, the standby logical unit searches the information on the applicable address. Then from the applicable information, it also searches information on the main memory area controlled by the main logical unit to establish in itself and forms a processing environment and state identical to the main logical unit such that the standby logical unit takes over all or a portion of the processing of the main logical unit. This enables the construction of a server system of high operability to overcome failures and poor response times by failover and cloning, etc.
摘要:
For reducing instruction cache misses in emulation by a dynamic conversion method, it is judged whether or not instruction cache conflict with a portion of an emulation program, whose execution frequency is high, occurs in an emulation operation, and only addresses not bringing about cache conflict are set as converted instruction storing areas. Thereby, instruction cache misses in a dynamic conversion type emulator are reduced, and the emulation speed is improved.
摘要:
A construction of the present invention includes a procedure of setting in advance a storing area in a converted instruction storing area table for recording a corresponding relation between a program before conversion and a storing address of a converted program at an initialization processing portion of an emulation program. In setting the storing area, address information on a memory on a portion whose execution frequency is high upon an emulation operation is acquired, and an address that brings about cache conflict on an instruction cache with the portion whose execution frequency is high is excepted and set as an area to store therein a converted instruction.
摘要:
There is provided with an information processing apparatus comprising a DRAM, a memory controller configured to access the DRAM, and a bus master configured to send, to the memory controller, an access request to the DRAM, the bus master comprises a transmission unit configured to transmit, to the memory controller, using a signal indicating a type of burst access which is requested of the memory controller by the bus master, an instruction to designate that an auto-precharge operation is not to be performed after accessing the first address, and an instruction to designate that an auto-precharge operation is to be performed after accessing the first address.
摘要:
There is provided with an information processing apparatus comprising a DRAM, a memory controller configured to access the DRAM, and a bus master configured to send, to the memory controller, an access request to the DRAM, the bus master comprises a transmission unit configured to transmit, to the memory controller, using a signal indicating a type of burst access which is requested of the memory controller by the bus master, an instruction to designate that an auto-precharge operation is not to be performed after accessing the first address, and an instruction to designate that an auto-precharge operation is to be performed after accessing the first address.
摘要:
According to an image processing apparatus and a control method thereof of the present invention, a graphics drawing result is obtained in accordance with one or more graphics drawing commands included in drawing commands, a clipping command is generated from one or more moving image drawing commands included in the drawing commands, and clipped graphics is obtained by clipping the graphics drawing result using the clipping command. Further, moving image data processed in accordance with the one or more moving image drawing commands included in the drawing commands is generated, and the generated moving image data and the clipped graphics are composed and output.
摘要:
In a display system which includes a plurality of space management units which are hierarchically connected, and each of which manages a predetermined space region as a management space region, a content management unit which manages one or a plurality of contents, and a plurality of display units which display information on display regions. Wherein each of the plurality of display units displays a content arranged in a subspace corresponding to a self display region in the management space region.
摘要:
This apparatus obtains a mapped position (xo, yo) of (xi, yi), acquires (xi′, yi′) by performing inverse mapping for each reference position being based on (xo, yo), obtains (xi″, yi″) by performing inverse mapping for integer parts xoc and yoc of xo and yo respectively, and performs interpolation using fraction parts of x- and y-coordinates of a position, of (xi″, yi″) and (xi′, yi′), whose integer parts of the x- and y-coordinates respectively coincide with xi and yi, and peripheral pixel values of (xi, yi).
摘要:
An information processing system for performing processing of dividing a moving image into tiles and packetizing and outputting information corresponding to each tile includes a process time measuring packet generation unit adapted to generate and transmit a process time measuring packet in which a packet sending time is set to measure a packet process time, a packet process time measuring unit adapted to measure, based on the packet sending time set in the process time measuring packet and the reception time of the process time measuring packet, the packet process time necessary for processing a packet, a determination unit adapted to determine, based on the packet process time, the timestamp of the moving image divided into the tiles, and a packetization unit adapted to execute processing of packetizing and outputting the timestamp and the information of the moving image divided into the tiles.