公开(公告)号：JP2012128628A

公开(公告)日：2012-07-05

申请号：JP2010279022

申请日：2010-12-15

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： NAKAIKE TAKUYA

IPC: G06F9/45

CPC classification number: G06F8/4443 ,  G06F8/443 ,  G06F8/4441 ,  G06F8/447

Abstract: PROBLEM TO BE SOLVED: To provide an apparatus and method in which a binary string composed of a plurality of instructions including a function call can be dynamically rewritten.SOLUTION: An apparatus is provided which includes a transactional memory that enables exclusive control of a transaction. The apparatus comprises a first code generator 200 which interprets a program and generates a first code in which a starting instruction to start a transaction and an ending instruction to end the transaction are inserted before and after an instruction string composed of a plurality of instructions for executing designated processing in the program, a second code generator 201 which generates a second code using a plurality of instructions in predetermined timing in response to the designated processing, and a code writer 202 which overwrites an instruction string of the first code with the second code or writes the second code into a part of the first code, within the transaction.

Abstract translation: 要解决的问题：提供一种装置和方法，其中可以动态地重写由包括功能调用的多个指令组成的二进制串。 解决方案：提供一种装置，其包括事务存储器，其实现事务的排他性控制。 该装置包括第一代码生成器200，其解释程序并生成其中开始交易的开始指令和结束交易的结束指令的第一代码被插入到由用于执行的多个指令组成的指令串之前和之后 程序中的指定处理，响应于指定处理在预定定时中使用多个指令生成第二代码的第二代码生成器201以及用第二代码或者第二代码覆盖第一代码的指令串的代码写入器202， 在事务中将第二个代码写入第一个代码的一部分。 版权所有（C）2012，JPO＆INPIT

公开(公告)号：JP2004213220A

公开(公告)日：2004-07-29

申请号：JP2002380356

申请日：2002-12-27

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： NAKAIKE TAKUYA ,  KONDO TAKESHI ,  KITAYAMA FUMIHIKO ,  HIROSE SHINICHI

IPC: G06F9/44

CPC classification number: G06F8/75

Abstract: PROBLEM TO BE SOLVED: To provide a method for automatically and properly dividing an application block realizing a mechanism dynamically creating a Web page in a server into several sections.
SOLUTION: This program editing device is provided with a user interface control part 10 receiving an input of a division object instruction designating a part to be divided from a processing objective JSP file, a JSP analysis part 20 deciding an accompanying division part to be divided with the part designated by the division object instruction, and a division execution part 30 dividing the part designated by the division object instruction and the accompanying division part from the JSP file. In the JSP analysis part 20, a variable in the JSP file and an object dependency relationship are analyzed, and a copyable part is divided by copying, so that the same operation as that by a processing objective program file is secured by the program file after division.
COPYRIGHT: (C)2004,JPO&NCIPI

Abstract translation: 要解决的问题：提供一种用于将实现将服务器中的网页动态创建的机制的应用程序块自动适当地划分为几个部分的方法。 解决方案：该程序编辑装置设置有用户接口控制部分10，接收从处理目标JSP文件指定要分割的部分的分割对象指令的输入，JSP分析部分20决定相应的分割部分 与由分割对象指令指定的部分分开，以及除法执行部30，其将由分割对象指令指定的部分和与JSP文件相关的分割部分分割。 在JSP分析部分20中，分析JSP文件中的变量和对象依赖关系，并通过复制来划分可复制部分，从而通过程序文件确保与处理目标程序文件相同的操作 师。 版权所有（C）2004，JPO＆NCIPI

公开(公告)号：JP2004046745A

公开(公告)日：2004-02-12

申请号：JP2002206202

申请日：2002-07-15

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： ABE MARI ,  KOYANAGI MITSUO ,  ONO KOICHI ,  HORI MASAHIRO ,  NAKAIKE TAKUYA

IPC: G06F17/21 ,  G06F15/00 ,  G06F17/22 ,  G06F17/30

CPC classification number: G06F17/2247 ,  G06F17/2211

Abstract: PROBLEM TO BE SOLVED: To continuously and properly indicate a desired element regardless of a change in a structured document when the prescribed element is indicated in the structured document. SOLUTION: An indication information generating system includes: a difference calculating part 20 for calculating a difference between the structured documents; and an X-path updating part 30 for generating indication information which indicates a corresponding part in the other structured document from indication information which indicates a prescribed part in the prescribed structured document, based on information concerning the difference calculated by the difference calculating part 20. COPYRIGHT: (C)2004,JPO

公开(公告)号：JP2011128975A

公开(公告)日：2011-06-30

申请号：JP2009288102

申请日：2009-12-18

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： NAKAIKE TAKUYA ,  MICHAEL MAGED M

IPC: G06F9/52

CPC classification number: G06F9/5038 ,  G06F9/528

Abstract: PROBLEM TO BE SOLVED: To provide an exclusive control method between tasks having a light processing load. SOLUTION: This system includes: a lock state storage unit 26 for storing update information whose value is updated in accordance with the matter that any task acquires and releases an exclusive lock, and task identification information for identifying a task acquiring the exclusive lock; an exclusive execution unit 28 for acquiring the exclusive lock to execute processing of a critical section included in a first task, releasing the exclusive lock after the processing of the critical section included in the first task, and also updating the update information; and a non-exclusive execution unit 30 for executing processing of a critical section included in a second task without acquiring the exclusive lock, and determining that the processing of the critical section is successful on the condition that update information is not changed before and after the processing of the critical section included in the second task. COPYRIGHT: (C)2011,JPO&INPIT

Abstract translation: 要解决的问题：提供具有光处理负载的任务之间的排他控制方法。 解决方案：该系统包括：锁定状态存储单元26，用于存储根据任务获取并释放排他锁的事件来更新其值的更新信息;以及用于识别获取排他锁的任务的任务识别信息 ; 排除执行单元28，用于获取独占锁以执行包括在第一任务中的关键部分的处理，在包括在第一任务中的关键部分的处理之后释放排他锁，并且还更新更新信息; 以及非排他性执行单元30，用于执行包括在第二任务中的关键部分的处理而不获取排他锁定，并且在更新信息在前后不改变的条件下确定关键部分的处理成功 包括在第二任务中的关键部分的处理。 版权所有（C）2011，JPO＆INPIT

公开(公告)号：JP2006127302A

公开(公告)日：2006-05-18

申请号：JP2004316996

申请日：2004-10-29

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： NAKAIKE TAKUYA ,  KOMATSU HIDEAKI ,  KONO SEIICHI

IPC: G06F9/45

CPC classification number: G06F1/3203 ,  G06F8/4432 ,  G06F9/4893 ,  Y02D10/24 ,  Y02D10/41

Abstract: PROBLEM TO BE SOLVED: To reduce energy consumption accompanying the execution of a program by adjusting the computing capacity of a central processing unit more accurately than before. SOLUTION: An information processor for setting appropriate computing capacity to the central processing unit capable of changing the computing capacity during execution comprises: an execution time measurement part for changing the computing capacity of the central processing unit and measuring execution time each time a program area is executed during the execution of an execution program for each of the plurality of program areas included in the execution program; an appropriate computing capacity judgement part for judging the appropriate computing capacity with which the program area is to be executed by the central processing unit on the basis of the execution time for each computing capacity measured for the respective program areas; and a computing capacity setting part for performing setting for executing the program area by the appropriate computing capacity of the program area when the program area is executed again during the execution of the execution program for the respective program areas. COPYRIGHT: (C)2006,JPO&NCIPI

Abstract translation: 要解决的问题：通过比以前更精确地调整中央处理单元的计算能力，减少伴随程序执行的能量消耗。 解决方案：用于向中央处理单元设置能够在执行期间改变计算能力的适当计算能力的信息处理器包括：执行时间测量部分，用于改变中央处理单元的计算能力，并且每次测量执行时间 在对执行程序中包括的多个程序区域中的每一个执行执行程序期间执行程序区域; 基于针对各个程序区域测量的每个计算能力的执行时间，判断中央处理单元执行程序区域的适当计算能力的适当计算能力判断部分; 以及计算能力设定部件，用于当在执行相应程序区域的执行程序期间再次执行程序区域时，执行程序区域的适当计算能力执行程序区域的设置。 版权所有（C）2006，JPO＆NCIPI

公开(公告)号：JP2011044078A

公开(公告)日：2011-03-03

申请号：JP2009193089

申请日：2009-08-24

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： TAKASU TOSHIHIKO ,  NAKAIKE TAKUYA

IPC: G06F11/14

CPC classification number: G06F9/3834 ,  G06F9/3842 ,  G06F9/3851 ,  G06F9/3861 ,  G06F11/1407 ,  G06F11/1479 ,  G06F11/1497 ,  G06F11/1629 ,  G06F11/1695

Abstract: PROBLEM TO BE SOLVED: To provide a system for recovering from transient fault. SOLUTION: A leading thread and a trailing thread are executed in parallel. Assuming that no transient fault occurs in each section, a system is speculatively executed in the section, with the leading thread and the trailing thread preferably being assigned to two different cores. At this time, the leading thread and the trailing thread are simultaneously executed, performing a buffering operation on a thread local area without performing a write operation on a shared memory. When the execution results of the two threads match each other, the content buffered to the thread local area is committed and written to the shared memory. When the execution results of the two threads do not match each other, the leading thread and the trailing thread are rolled back to a preceding commit point and re-executed. Matching of the execution results of the two threads can be guaranteed, to prevent an error by transient fault. COPYRIGHT: (C)2011,JPO&INPIT

Abstract translation: 要解决的问题：提供一种从瞬态故障中恢复的系统。

解决方案：并行执行前导线程和后退线程。 假设每个部分没有发生瞬态故障，则在该部分中推测性地执行一个系统，其中前导线和尾线最好分配给两个不同的核。 此时，同时执行前导线程和后退线程，对线程本地区域执行缓冲操作，而不对共享存储器执行写入操作。 当两个线程的执行结果相互匹配时，缓存到线程局部区域的内容被提交并写入共享存储器。 当两个线程的执行结果彼此不匹配时，前导线程和后退线程将回滚到先前的提交点并重新执行。 可以保证两个线程执行结果的匹配，以防止短暂故障发生错误。 版权所有（C）2011，JPO＆INPIT

公开(公告)号：JP2010117869A

公开(公告)日：2010-05-27

申请号：JP2008290387

申请日：2008-11-12

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： NAKAIKE TAKUYA ,  OHIRA REI ,  OGASAWARA TAKESHI

IPC: G06F12/00

CPC classification number: G06F9/466

Abstract: PROBLEM TO BE SOLVED: To improve execution efficiency of a program, in relation to a memory control device, the program and a method.
SOLUTION: The memory control device controls access to a shared memory by a transaction unit and includes: a management unit for storing a version of data stored in the shared memory; a log storage unit which, in response to execution of update processing for a shared memory under processing of a transaction, stores, for each transaction, update entry including a version of data subjected to the update processing; and a control unit which, in response to a commit request provided for a transaction, if a version included in the update entry matches the version stored in the management unit, a processing result is written according to the execution of the update processing into the shared memory; and if they do not match, the update processing is re-run and a rerun result is written to the shared memory.
COPYRIGHT: (C)2010,JPO&INPIT

Abstract translation: 要解决的问题：为了提高程序的执行效率，相对于存储器控制装置，程序和方法。 存储器控制装置控制由交易单元对共享存储器的访问，并且包括：管理单元，用于存储存储在共享存储器中的数据的版本; 日志存储单元，响应于在处理事务处理对共享存储器的更新处理的执行，针对每个事务存储包括经过更新处理的数据的版本的更新条目; 以及控制单元，响应于为交易提供的提交请求，如果包括在更新条目中的版本与存储在管理单元中的版本相匹配，则将处理结果根据更新处理的执行写入共享 记忆; 并且如果它们不匹配，则重新运行更新处理并且将重新运行结果写入共享存储器。 版权所有（C）2010，JPO＆INPIT

公开(公告)号：JP2006155525A

公开(公告)日：2006-06-15

申请号：JP2004349008

申请日：2004-12-01

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： NAKAIKE TAKUYA ,  KOMATSU HIDEAKI

IPC: G06F9/48 ,  B25B21/00 ,  B25B21/02 ,  B25D9/00 ,  G06F1/32 ,  G06F11/30

CPC classification number: B25B21/00 ,  B25B21/026 ,  H04L67/02 ,  H04L67/322

Abstract: PROBLEM TO BE SOLVED: To bring response time such as that of a web server close to a target value by more few overheads than up to now. SOLUTION: This control unit is a control unit 25 which controls average response time that is from the time when the information processing system receives a processing request until the time when the system replies to the processing request; and the control unit provides a control unit which is provided with a target response time acquiring part 200 which acquires a target response time that is a target value of an average response time, a predicted response time calculation part 230 which calculates a predicted response time that is a predicted value of average response time at the time when standard period set beforehand passes after the operation mode is set to the information processing system at the case in which either of a plurality of operation modes whose processing performance are different each other, and an operation mode setting part 240 which sets the operation mode to the information processing system when a predicted response time calculated by the predicted response time calculation part is under the target response time. COPYRIGHT: (C)2006,JPO&NCIPI

Abstract translation: 要解决的问题：通过比现在更多的开销，使Web服务器的响应时间接近目标值。 解决方案：该控制单元是控制从信息处理系统接收到处理请求直到系统回复处理请求的时间的平均响应时间的控制单元25。 并且控制单元提供控制单元，该控制单元设置有获取作为平均响应时间的目标值的目标响应时间的目标响应时间获取部200，计算预测响应时间计算部230， 是在其处理性能彼此不同的多种操作模式中的任一种的情况下，在将操作模式设置为信息处理系统之后，预先设定的标准周期的时间的平均响应时间的预测值， 操作模式设定部240，其在由所述预测响应时间计算部计算出的预测响应时间在所述目标响应时间以下时，将操作模式设定为所述信息处理系统。 版权所有（C）2006，JPO＆NCIPI

公开(公告)号：AU2021285952A1

公开(公告)日：2022-10-27

申请号：AU2021285952

申请日：2021-05-14

Applicant: IBM

Inventor： ZHANG QI ,  NOVOTNY PETR ,  MIN HONG ,  NAIR RAVI ,  RAMJI SHYAM ,  YU LEI ,  NAKAIKE TAKUYA ,  KAWAHITO MOTOHIRO

IPC: G06N99/00 ,  G06N3/04

Abstract: Techniques for refinement of data pipelines are provided. An original file of serialized objects is received, and an original pipeline comprising a plurality of transformations is identified based on the original file. A first computing cost is determined for a first transformation of the plurality of transformations. The first transformation is modified using a predefined optimization, and a second cost of the modified first transformation is determined. Upon determining that the second cost is lower than the first cost, the first transformation is replaced, in the original pipeline, with the optimized first transformation.

公开(公告)号：DE112021001767T5

公开(公告)日：2023-01-12

申请号：DE112021001767

申请日：2021-05-14

Applicant: IBM

Inventor： ZHANG QI ,  NOVOTNY PETR ,  MIN HONG ,  NAIR RAVI ,  RAMJI SHYAM ,  YU LEI ,  NAKAIKE TAKUYA ,  KAWAHITO MOTOHIRO

IPC: G06N20/00

Abstract: Bereitgestellt werden Methoden zur Verfeinerung von Daten-Pipelines. Eine ursprüngliche Datei von serialisierten Objekten wird empfangen, und eine ursprüngliche Pipeline, die eine Mehrzahl von Transformationen aufweist, wird auf Grundlage der ursprünglichen Datei identifiziert. Für eine erste Transformation der Mehrzahl von Transformationen werden erste Datenverarbeitungskosten ermittelt. Die erste Transformation wird unter Verwendung einer vordefinierten Optimierung modifiziert, und zweite Kosten der modifizierten ersten Transformation werden ermittelt. Wenn ermittelt wird, dass die zweiten Kosten niedriger als die ersten Kosten sind, wird die erste Transformation in der ursprünglichen Pipeline durch die optimierte erste Transformation ersetzt.