Method of processing object in computer, program and system
    11.
    发明专利
    Method of processing object in computer, program and system 有权
    在计算机,程序和系统中处理对象的方法

    公开(公告)号:JP2012104038A

    公开(公告)日:2012-05-31

    申请号:JP2010253987

    申请日:2010-11-12

    CPC classification number: G06F9/4428 G06F9/45504

    Abstract: PROBLEM TO BE SOLVED: To fasten a heap objecting process.SOLUTION: In a processing system, an address A of a heap-objected stack object is added to a working set and a stack is scanned. Next, a stack frame to allocate the address A is detected and defined as a current stack frame. When a parameter indicating the address added to the working set is found in the current stack frame, that address is rewritten into an address of a corresponding heap object. When a field of the stack object in the current stack frame indicates the address included in the working set, in the processing system, the address of that stack object is added to the working set. If the stack frame viewed at present includes a pointer for the address included in the working set, processing is advanced to a next stack frame but if not, in the processing system, scanning of the stack is finished there.

    Abstract translation: 要解决的问题:要紧固堆对象进程。 解决方案:在处理系统中,将堆对象堆栈对象的地址A添加到工作集中,并扫描堆栈。 接下来,检测分配地址A的堆栈帧并将其定义为当前堆栈帧。 当在当前堆栈帧中找到指示添加到工作集的地址的参数时,将该地址重写为相应堆对象的地址。 当当前堆栈帧中的堆栈对象的字段指示包含在工作集中的地址时,在处理系统中,将该堆栈对象的地址添加到工作集中。 如果当前查看的堆栈帧包括用于包括在工作集中的地址的指针,则处理进行到下一堆栈帧,但是如果不是,则在处理系统中,堆栈的扫描在那里完成。 版权所有(C)2012,JPO&INPIT

    IMAGE DISPLAY UNIT
    12.
    发明专利

    公开(公告)号:JPH06250618A

    公开(公告)日:1994-09-09

    申请号:JP2396393

    申请日:1993-02-12

    Applicant: IBM

    Abstract: PURPOSE: To surely detect an object instructed by an operator. CONSTITUTION: By image data from an image data memory 76 and data on the number and position of the object to the images outputted by an object control part 68, a liquid crystal unit 14 is driven through a driver 74 and a picture composed of the images and the objects is formed on the screen. When the operator processes an instruction switch 36, signals including the brightness and darkness state of the object from an instruction device 12 are inputted through the SW judgement part 62 of a PC 18 to a brightness and darkness judgement part 64, brightness and darkness judgement is started in the brightness and darkness judgement part 64 and the brightness of the respective objects is made to fluctuate by the object control part 68. Along with the brightness fluctuation of the objects, the instructed object is decided based on the brightness detected by a photosensor in the brightness and darkness judgement part 64 and a processing (screen changeover), corresponding to the decided object is executed by an object corresponding processing part 66.

    MULTIPROCESSOR SYSTEM AND ITS MESSAGE TRANSMISSION AND RECEPTION CONTROLLER

    公开(公告)号:JPH04312160A

    公开(公告)日:1992-11-04

    申请号:JP8900491

    申请日:1991-03-29

    Applicant: IBM

    Abstract: PURPOSE: To provide a generalized aperiodic message transmission system by making each processor generate a response signal to a message passing request signal from another processor, counting response signals, and sending a message to processors as many as the counted value. CONSTITUTION: A DCR 5 specifies the address of the message and when message transmission is successful, the contents of an MDR 6 flow to a message data bus 1b and reaches the destination processor. An LNR 7 specifies whether or not a minimum number of processors are ready to receive the message and an HNR 8 determines whether or not the message can be sent to a maximum number of processors. A parallel counter 9 counts how many processors return affirmative responses and a parallel comparator 10 compares the contents of the parallel counter 9 with the contents of the LNR 7 and selects processors as many as the number of the LNR 8 when affirmative responses are returned from more processors than the value set in the LNR 7, thereby setting them to the DCR 5.

    Method, program and system for controlling operation of computer
    14.
    发明专利
    Method, program and system for controlling operation of computer 有权
    用于控制计算机操作的方法,程序和系统

    公开(公告)号:JP2012150716A

    公开(公告)日:2012-08-09

    申请号:JP2011009968

    申请日:2011-01-20

    CPC classification number: G06F9/52

    Abstract: PROBLEM TO BE SOLVED: To provide processing for appropriately making sure of such a case that stack allocation is possible when the processing of fork() or join() is performed in the framework of Fork-Join.SOLUTION: In the framework of Fork-Join, following determination processing is implemented. In processing of inputting a task object to the queue of work stealing, it is determined that the task object does not escape. This processing is performed by, for example, fork(). In processing of waiting for the end of the task object input to the queue, it is determined that the task object does not escape. This processing is performed by, for example, join(). When the task object escapes in any case other than the above mentioned case, it is determined that the task object does not escape. When it is determined that the task object does not escape, the stack allocation of the task object is performed.

    Abstract translation: 要解决的问题:为了适当地确定在fork()或join()在Fork-Join的框架中执行的情况下可以进行堆栈分配的情况的处理。 解决方案:在Fork-Join的框架下,执行以下确定处理。 在将任务对象输入到工作队列中的处理中,确定任务对象不能逃脱。 该处理由例如fork()来执行。 在处理等待输入到队列的任务对象的结束时,确定任务对象不会转义。 该处理通过例如join()执行。 当任务对象在上述情况以外的任何情况下转义时,确定任务对象不能转义。 当确定任务对象不转义时,执行任务对象的堆栈分配。 版权所有(C)2012,JPO&INPIT

    Code generation method, system, and program
    15.
    发明专利
    Code generation method, system, and program 有权
    代码生成方法,系统和程序

    公开(公告)号:JP2012053566A

    公开(公告)日:2012-03-15

    申请号:JP2010194224

    申请日:2010-08-31

    CPC classification number: G06F8/4443 G06F9/45516

    Abstract: PROBLEM TO BE SOLVED: To efficiently achieve in-lining for code optimization.SOLUTION: When a method is a supposition multimethod, a distribution of types in execution of a group of actual arguments is inspected by type inference and profiling. If appearance of the group of actual arguments deviates, the method is specialized for the group of types of high frequency, and the method and the group of types are cached as a key. A code at a call source is optimized so as to call the specialized method by performing run-time inspection for several groups of types of high frequency. When the code of the specialized method is small enough, in-lining is achieved.

    Abstract translation: 要解决的问题:有效地实现代码优化的内联。

    解决方案:当一种方法是假设多方法时,通过类型推断和分析来检查实际参数组执行中的类型分布。 如果实际参数组的出现偏差,则该方法专门针对高频类型的组,并且方法和类型组被缓存为关键。 优化呼叫源上的代码,以便通过对几组高频类型执行运行时检查来调用专门的方法。 当专门方法的代码足够小时,可以实现内衬。 版权所有(C)2012,JPO&INPIT

    Compiler program, compilation method and computer system
    17.
    发明专利
    Compiler program, compilation method and computer system 有权
    编译程序,编译方法和计算机系统

    公开(公告)号:JP2011065220A

    公开(公告)日:2011-03-31

    申请号:JP2009212881

    申请日:2009-09-15

    CPC classification number: G06F8/4443 G06F8/433

    Abstract: PROBLEM TO BE SOLVED: To improve the performance of an execution code by applying delay evaluation to imperative language having side effects such as Java (R) or PHP. SOLUTION: The compilation method includes: generating a dependence graph for data and side effects in the analysis of a source code; identifying a portion of the program able to be delayed based on a postdominator tree; generating a code for performing delay; profiling a location where the delay has been forced by a profiler during execution; in-lining a delay closure code determined to be forced with high frequency based on the profile into a location where the delay has been forced; generating a code by applying partial evaluation; and eliminating an intermediate data structure by the partial evaluation in the generated code. COPYRIGHT: (C)2011,JPO&INPIT

    Abstract translation: 要解决的问题:通过对具有副作用的命令式语言(如Java(R)或PHP))应用延迟评估来提高执行代码的性能。

    解决方案:编译方法包括:在源代码分析中生成数据和副作用的依赖图; 基于一个后缀树识别能够被延迟的程序的一部分; 生成用于执行延迟的代码; 分析在执行期间由分析器强制延迟的位置; 将基于简档强制为高频的延迟关闭码内插到已经被强制延迟的位置; 通过应用部分评估生成代码; 并通过在生成的代码中的部分评估来消除中间数据结构。 版权所有(C)2011,JPO&INPIT

    Method for optimizing processing of character string upon execution of program, computer system of the same, and computer program
    18.
    发明专利
    Method for optimizing processing of character string upon execution of program, computer system of the same, and computer program 审中-公开
    方案执行程序,计算机系统及计算机程序优化处理方法

    公开(公告)号:JP2010277440A

    公开(公告)日:2010-12-09

    申请号:JP2009131016

    申请日:2009-05-29

    CPC classification number: G06F8/443 G06F8/4434 G06F8/4441

    Abstract: PROBLEM TO BE SOLVED: To provide a mechanism of creating a program with high execution efficiency without deteriorating the productivity of the program. SOLUTION: In a computer system that executes the program including the processing of a character string, a method for optimizing the processing of the character string upon the execution of the program by using characteristic information showing the characteristic of the character string associated with the character string is provided. The method includes a step of determining from the characteristic of a first character string and arithmetic processing regarding the first character string, the character string characteristic information for at least one of the first character string and a second character string which is a result of the arithmetic processing, and a step of making the determined characteristic information associated with at least one character string. Furthermore, a computer system and a computer program for executing the program including the processing of the character string and optimizing the processing of the character string upon execution of the program using the characteristic information are provided. COPYRIGHT: (C)2011,JPO&INPIT

    Abstract translation: 要解决的问题:提供一种创建具有高执行效率的程序的机制,而不会降低程序的生产率。 解决方案:在执行包括处理字符串的程序的计算机系统中,通过使用表示与...相关联的字符串的特性的特征信息来优化程序执行时的字符串的处理的方法 提供字符串。 该方法包括从第一字符串的特性和关于第一字符串的算术处理确定作为算术结果的第一字符串和第二字符串中的至少一个的字符串特征信息的步骤 处理和与至少一个字符串相关联的所确定的特征信息的步骤。 此外,提供了一种用于执行程序的计算机系统和计算机程序,包括处理字符串,并且在使用特征信息执行程序时优化字符串的处理。 版权所有(C)2011,JPO&INPIT

    Caching mechanism based on update frequency
    19.
    发明专利
    Caching mechanism based on update frequency 有权
    基于更新频率的缓存机制

    公开(公告)号:JP2010092222A

    公开(公告)日:2010-04-22

    申请号:JP2008260892

    申请日:2008-10-07

    Abstract: PROBLEM TO BE SOLVED: To provide a cache configuration for effectively using a cache in Map-Reduce processing on the basis of an update frequency.
    SOLUTION: A method for constructing a cache mechanism in a Map-Reduce processing system performing distributed processing for a plurality of items of data includes: a step of classifying the plurality of items of data to be processed into a plurality of groups on the basis of each update frequency of the data; a step of calculating a group update frequency that is each update frequency of the plurality of groups on the basis of data update frequency that is the update frequency of the data each constituting the plurality of groups; a step of generating a partial result of a Map-Reduce processing stage for the groups of which the group update frequency is not larger than a threshold among the plurality of groups; and a step of caching the generated partial result.
    COPYRIGHT: (C)2010,JPO&INPIT

    Abstract translation: 要解决的问题:提供基于更新频率在Map-Reduce处理中有效地使用高速缓存的高速缓存配置。 解决方案:一种在对多个数据项执行分布式处理的Map-Reduce处理系统中构建高速缓存机制的方法包括:将要处理的多个数据项目分类为多个组的步骤, 数据的每个更新频率的基础; 基于作为构成所述多个组的数据的更新频率的数据更新频率来计算作为所述多个组的每个更新频率的组更新频率的步骤; 生成所述组更新频率不大于所述组中的阈值的组的映射减少处理级的部分结果的步骤; 以及缓存生成的部分结果的步骤。 版权所有(C)2010,JPO&INPIT

    Truth/falsehood discrimination system for object in metaverse, method, and computer program therefor
    20.
    发明专利
    Truth/falsehood discrimination system for object in metaverse, method, and computer program therefor 有权
    用于对象的真实/虚假的辨别系统,方法和计算机程序

    公开(公告)号:JP2009217387A

    公开(公告)日:2009-09-24

    申请号:JP2008058489

    申请日:2008-03-07

    Abstract: PROBLEM TO BE SOLVED: To provide a means which can discriminate truth/falsehood of an object with a simple system configuration assuming free creation of objects by a user in a metaverse and introduction of objects from outside the metaverse. SOLUTION: A server computer 1 includes a storage section for associating an object ID for identifying an object with authenticity information indicating the authenticity of the object, regarding the object, and storing them as object information, a communication section for communicating with respective client computers 3A to 3N, and an inquiry section which responds to reception of an inquiry request for an object including a corresponding object ID from a client computer, and allows the communication section to transmit authenticity information to the client computer on condition that the authenticity information corresponding to the object ID is stored in the storage section. COPYRIGHT: (C)2009,JPO&INPIT

    Abstract translation: 要解决的问题:提供一种能够以简单的系统配置来区分对象的真实/虚假的手段,假设用户在虚拟人中自由创建对象并从外部引入对象。 解决方案:服务器计算机1包括存储部分,用于将用于识别对象的对象ID与指示对象的真实性的真实性信息相关联,并将其作为对象信息存储;通信部分,用于与对象通信 客户端计算机3A至3N,以及查询部分,其响应于从客户端计算机接收到包括相应对象ID的对象的查询请求,并且允许通信部分将真实性信息发送到客户端计算机,条件是真实性信息 对应于对象ID被存储在存储部分中。 版权所有(C)2009,JPO&INPIT

Patent Agency Ranking