专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明申请

US20110314225A1 COMPUTATIONAL RESOURCE ASSIGNMENT DEVICE, COMPUTATIONAL RESOURCE ASSIGNMENT METHOD AND COMPUTATIONAL RESOURCE ASSIGNMENT PROGRAM 有权
标题翻译：计算资源分配设备，计算资源分配方法和计算资源分配方案
公开(公告)号：US20110314225A1
公开(公告)日：2011-12-22
申请号：US13138354
申请日：2010-02-12
申请人： Kosuke Nishihara , Kazuhisa Ishizaka
发明人： Kosuke Nishihara , Kazuhisa Ishizaka
IPC分类号： G06F12/08
CPC分类号： G06F9/5088
摘要： In a multi-core processor system, cache memories are provided respectively for a plurality of processors. An assignment management unit manages assignment of tasks to the processors. A cache status calculation unit calculates a cache usage status such as a memory access count and a cache hit ratio, with respect to each task. A first processor handles a plurality of first tasks that belong to a first process. If computation amount of the first process exceeds a predetermined threshold value, the assignment management unit refers to the cache usage status to preferentially select, as a migration target task, one of the plurality of first tasks whose memory access count is smaller or whose cache hit ratio is higher. Then, the assignment management unit newly assigns the migration target task to a second processor handling another process different from the first processor.
摘要翻译：在多核处理器系统中，分别为多个处理器提供高速缓存存储器。分配管理单元管理任务到处理器的分配。高速缓存状态计算单元针对每个任务计算诸如存储器访问计数和高速缓存命中率的高速缓存使用状态。第一处理器处理属于第一进程的多个第一任务。如果第一处理的计算量超过预定阈值，则分配管理单元参照高速缓存使用状态，优先选择存储器访问次数较小的多个第一任务中的一个，或者其缓存命中比例较高。然后，分配管理单元将迁移目标任务新分配给处理与第一处理器不同的其他处理的第二处理器。

2. 发明申请

US20110276980A1 COMPUTING RESOURCE ALLOCATION DEVICE, COMPUTING RESOURCE ALLOCATION SYSTEM, COMPUTING RESOURCE ALLOCATION METHOD THEREOF AND PROGRAM 有权
标题翻译：计算资源分配设备，计算资源分配系统，计算资源分配方法及程序
公开(公告)号：US20110276980A1
公开(公告)日：2011-11-10
申请号：US13145900
申请日：2010-02-16
申请人： Kosuke Nishihara , Kazuhisa Ishizaka
发明人： Kosuke Nishihara , Kazuhisa Ishizaka
IPC分类号： G06F9/50
CPC分类号： B60W40/02
摘要： Provided is a computing resource allocation device capable of allocating computing resources to accommodate changing activity patterns. The device is equipped with an external environment recognition means that analyzes input values from sensors to specify the current environment, a memory means that stores a table in which the sensors required to specify the environment are correlated, a transition frequency computation means that computes the transition frequency at which a transition is made from an environment to another environment, and a computing resource allocation means that computes the amount of allocation of the computing resources to be used for the analysis based on the current environment by referencing the table and the transition frequency, and that allocates the computing resources for the analysis.
摘要翻译：提供了能够分配计算资源以适应变化的活动模式的计算资源分配装置。该装置配备有外部环境识别装置，其分析来自传感器的输入值以指定当前环境;存储装置，其存储指定环境所需的传感器相关联的表;计算转换的转换频率计算装置从环境向另一环境进行转换的频率;以及计算资源分配装置，其通过参考表和转换频率，根据当前环境计算要用于分析的计算资源的分配量，并分配用于分析的计算资源。

3. 发明授权

US08897372B2 Task allocation device, task allocation method, and storage medium storing task allocation program 有权
标题翻译：任务分配设备，任务分配方法和存储任务分配程序的存储介质
公开(公告)号：US08897372B2
公开(公告)日：2014-11-25
申请号：US13201150
申请日：2010-01-22
申请人： Kosuke Nishihara , Kazuhisa Ishizaka
发明人： Kosuke Nishihara , Kazuhisa Ishizaka
IPC分类号： G06F9/46 , H04N7/26 , G06F9/50 , G06F9/48
CPC分类号： G06F9/5027 , G06F9/4881 , G06F2209/483 , G06F2209/501
摘要： To reduce performance degradation due to a high-performance core's waiting for a processing result of a low-performance core in a multi-core processor including a plurality of cores with different running performance, included are a task pool that stores executable tasks for each computational core, a task scheduler, and a reference count analysis module that acquires hint information and a reference count estimation method from a running task and estimates a reference count of a task to be newly inserted into the task pool or a task existing in the task pool based on the specified method. The scheduler performs insertion and acquisition of a task by mainly using performance of the cores and the reference count.
摘要翻译：为了降低由于高性能核心等待包括具有不同运行性能的多个核心的多核处理器的低性能核心的处理结果而导致的性能下降，包括存储每个计算的可执行任务的任务池核心，任务调度器和参考计数分析模块，其从正在运行的任务中获取提示信息和参考计数估计方法，并且估计新插入任务池的任务的引用计数或任务池中存在的任务基于指定的方法。调度器通过主要使用内核的性能和引用计数来执行任务的插入和获取。

4. 发明申请

US20120324469A1 RESOURCE ALLOCATION APPARATUS, RESOURCE ALLOCATION METHOD, AND COMPUTER READABLE MEDIUM 有权
标题翻译：资源分配设备，资源分配方法和计算机可读介质
公开(公告)号：US20120324469A1
公开(公告)日：2012-12-20
申请号：US13581775
申请日：2010-12-09
申请人： Kosuke Nishihara , Kazuhisa Ishizaka
发明人： Kosuke Nishihara , Kazuhisa Ishizaka
IPC分类号： G06F9/50
CPC分类号： G06F9/50
摘要： A parameter determination unit 110 substitutes, for each of a plurality of applications, a recommended amount of resources and a quality of experience corresponding to the recommended amount of resources, and a minimum amount of resources and a quality of experience corresponding to the minimum amount of resources into a quality function fin expression (1) indicating a relation between an amount of resources R and a quality of experience Q, to determine parameters a and b. A resource amount determination unit 120 determines an amount of resources to be allocated to the plurality of applications using the quality function f for each application in which the parameters a and b are determined. The quality function f(x) is a monotonically increasing function having an inverse function f−1, connects (−∞) and (+∞), and is symmetrical with respect to x=0. Q=f(x)=f((R−a)/b) (1).
摘要翻译：参数确定单元110针对多个应用中的每一个替代与推荐的资源量相对应的推荐资源量和经验品质，以及最小量的资源和与最小量的资源转换成质量函数fin表达式（1），表示资源量R与体验质量Q之间的关系，以确定参数a和b。资源量确定单元120使用对确定参数a和b的每个应用的质量函数f来确定要分配给多个应用的资源量。质量函数f（x）是具有反函数f-1的单调递增函数，连接（-∞）和（+∞），并且相对于x = 0是对称的。 Q = f（x）= f（（R-a）/ b）（1）。

5. 发明申请

US20110047556A1 SYNCHRONIZATION CONTROL METHOD AND INFORMATION PROCESSING DEVICE 有权
标题翻译：同步控制方法和信息处理设备
公开(公告)号：US20110047556A1
公开(公告)日：2011-02-24
申请号：US12812936
申请日：2009-01-14
申请人： Kosuke Nishihara
发明人： Kosuke Nishihara
IPC分类号： G06F9/46
CPC分类号： G06F9/52
摘要： Provided are a synchronization control section that executes a current thread and a reference thread in parallel, a waiting time calculation section that calculates the time needed for the reference thread to reach a second synchronization point as a waiting time of the current thread when the reference thread does not reach the second synchronization point at a time when the current thread reaches a first synchronization point, a quality difference calculation section that estimates a quality difference between data that the current thread generates by referring to processing data at the second synchronization point of the reference thread and data that the current thread generates without referring to the processing data, and a synchronization determination section that determines whether to make the current thread wait until the reference thread reaches the second synchronization point depending on the waiting time and the magnitude of the quality difference.
摘要翻译：提供一种并行执行当前线程和参考线程的同步控制部分，等待时间计算部分，当等待时间计算部分将参考线程到达第二同步点所需的时间作为当前线程的等待时间时，在当前线程达到第一同步点的时刻没有到达第二同步点，质量差计算部分通过参考参考的第二同步点处的处理数据来估计当前线程生成的数据之间的质量差异线程和当前线程生成的数据，而不参考处理数据;以及同步确定部，其根据等待时间和质量差的大小来确定是否使当前线程等待直到参考线程到达第二同步点。

6. 发明申请

US20100246673A1 DYNAMIC IMAGE RECEIVING APPARATUS, DYNAMIC IMAGE RECEIVING METHOD AND PROGRAM 有权
标题翻译：动态图像接收装置，动态图像接收方法和程序
公开(公告)号：US20100246673A1
公开(公告)日：2010-09-30
申请号：US12680491
申请日：2008-09-19
申请人： Katsunori Tanaka , Atsushi Hatabu , Yuzo Senda , Katsutoshi SEKI , Tomoyoshi Kobori , Kosuke Nishihara , Soji Mori
发明人： Katsunori Tanaka , Atsushi Hatabu , Yuzo Senda , Katsutoshi SEKI , Tomoyoshi Kobori , Kosuke Nishihara , Soji Mori
IPC分类号： H04N7/12
CPC分类号： H04N21/4307 , H04N19/44
摘要： To provide a dynamic image receiving apparatus which receives dynamic image streams coded with inter-frame prediction such as MPEG from a plurality of channels, and collects the dynamic image streams containing intra-frame coded pictures from each channel in a short time. The dynamic image receiving apparatus includes: a time information accumulative processing device which accumulates code receiving time of the intra-frame coded picture of the dynamic image stream, and periodicity time information containing one of or both of presentation time information and decoding time information contained in the dynamic image stream for each dynamic image stream of the plurality of channels; code receiving time predicting devices which predict the code receiving time of the intra-frame coded pictures based on the periodicity time information; and a channel selection control device which controls channel selection of the dynamic image stream to be received based on the predicted code receiving time information.
摘要翻译：提供一种动态图像接收装置，其接收从多个频道接收诸如MPEG之类的帧间预测编码的动态图像流，并且在短时间内收集每个信道中包含帧内编码图像的动态图像流。动态图像接收装置包括：时间信息累积处理装置，其累积动态图像流的帧内编码图像的代码接收时间，以及包含显示时间信息和解码时间信息中的一个或两个的周期性时间信息针对所述多个频道的每个动态图像流的所述动态图像流; 代码接收时间预测装置，其基于周期时间信息来预测帧内编码图像的代码接收时间; 以及频道选择控制装置，其基于预测的代码接收时间信息来控制要接收的动态图像流的频道选择。

7. 发明授权

US08635405B2 Computational resource assignment device, computational resource assignment method and computational resource assignment program 有权
标题翻译：计算资源分配装置，计算资源分配方法和计算资源分配方案
公开(公告)号：US08635405B2
公开(公告)日：2014-01-21
申请号：US13138354
申请日：2010-02-12
申请人： Kosuke Nishihara , Kazuhisa Ishizaka
发明人： Kosuke Nishihara , Kazuhisa Ishizaka
IPC分类号： G06F13/00 , G06F13/28
CPC分类号： G06F9/5088
摘要： In a multi-core processor system, cache memories are provided respectively for a plurality of processors. An assignment management unit manages assignment of tasks to the processors. A cache status calculation unit calculates a cache usage status such as a memory access count and a cache hit ratio, with respect to each task. A first processor handles a plurality of first tasks that belong to a first process. If computation amount of the first process exceeds a predetermined threshold value, the assignment management unit refers to the cache usage status to preferentially select, as a migration target task, one of the plurality of first tasks whose memory access count is smaller or whose cache hit ratio is higher. Then, the assignment management unit newly assigns the migration target task to a second processor handling another process different from the first processor.
摘要翻译：在多核处理器系统中，分别为多个处理器提供高速缓存存储器。分配管理单元管理任务到处理器的分配。高速缓存状态计算单元针对每个任务计算诸如存储器访问计数和高速缓存命中率的高速缓存使用状态。第一处理器处理属于第一进程的多个第一任务。如果第一处理的计算量超过预定阈值，则分配管理单元参照高速缓存使用状态，优先选择存储器访问次数较小的多个第一任务中的一个，或者其缓存命中比例较高。然后，分配管理单元将迁移目标任务新分配给处理与第一处理器不同的其他处理的第二处理器。

8. 发明授权

US08555291B2 Synchronization control method and information processing device 有权
标题翻译：同步控制方法和信息处理装置
公开(公告)号：US08555291B2
公开(公告)日：2013-10-08
申请号：US12812936
申请日：2009-01-14
申请人： Kosuke Nishihara
发明人： Kosuke Nishihara
IPC分类号： G06F9/46 , H04B1/66
CPC分类号： G06F9/52
摘要： Provided are a synchronization control section that executes a current thread and a reference thread in parallel, a waiting time calculation section that calculates the time needed for the reference thread to reach a second synchronization point as a waiting time of the current thread when the reference thread does not reach the second synchronization point at a time when the current thread reaches a first synchronization point, a quality difference calculation section that estimates a quality difference between data that the current thread generates by referring to processing data at the second synchronization point of the reference thread and data that the current thread generates without referring to the processing data, and a synchronization determination section that determines whether to make the current thread wait until the reference thread reaches the second synchronization point depending on the waiting time and the magnitude of the quality difference.
摘要翻译：提供一种并行执行当前线程和参考线程的同步控制部分，等待时间计算部分，当等待时间计算部分将参考线程到达第二同步点所需的时间作为当前线程的等待时间时，在当前线程达到第一同步点的时刻没有到达第二同步点，质量差计算部分通过参考参考的第二同步点处的处理数据来估计当前线程生成的数据之间的质量差异线程和当前线程生成的数据，而不参考处理数据;以及同步确定部，其根据等待时间和质量差的大小来确定是否使当前线程等待直到参考线程到达第二同步点。

9. 发明授权

US08533726B2 Device, system, method, and program for computing resource allocation based on external environment and transition frequency 有权
标题翻译：基于外部环境和转换频率计算资源分配的设备，系统，方法和程序
公开(公告)号：US08533726B2
公开(公告)日：2013-09-10
申请号：US13145900
申请日：2010-02-16
申请人： Kosuke Nishihara , Kazuhisa Ishizaka
发明人： Kosuke Nishihara , Kazuhisa Ishizaka
IPC分类号： G06F9/46
CPC分类号： B60W40/02
摘要： Provided is a computing resource allocation device capable of allocating computing resources to accommodate changing activity patterns. The device is equipped with an external environment recognition means that analyzes input values from sensors to specify the current environment, a memory means that stores a table in which the sensors required to specify the environment are correlated, a transition frequency computation means that computes the transition frequency at which a transition is made from an environment to another environment, and a computing resource allocation means that computes the amount of allocation of the computing resources to be used for the analysis based on the current environment by referencing the table and the transition frequency, and that allocates the computing resources for the analysis.
摘要翻译：提供了能够分配计算资源以适应变化的活动模式的计算资源分配装置。该装置配备有外部环境识别装置，其分析来自传感器的输入值以指定当前环境;存储装置，其存储指定环境所需的传感器相关联的表;计算转换的转换频率计算装置从环境向另一环境进行转换的频率;以及计算资源分配装置，其通过参考表和转换频率，根据当前环境计算要用于分析的计算资源的分配量，并分配用于分析的计算资源。

10. 发明授权

US08447933B2 Memory access control system, memory access control method, and program thereof 有权
标题翻译：存储器访问控制系统，存储器访问控制方法及其程序
公开(公告)号：US08447933B2
公开(公告)日：2013-05-21
申请号：US12526233
申请日：2008-02-04
申请人： Kosuke Nishihara
发明人： Kosuke Nishihara
IPC分类号： G66F12/00
CPC分类号： G06F12/0862 , G06F12/0875 , H04N19/433 , H04N19/436 , H04N19/61
摘要： In a multi-core processor of a shared-memory type, deterioration in the data processing capability caused by competitions of memory accesses from a plurality of processors is suppressed effectively. In a memory access controlling system for controlling accesses to a cache memory in a data read-ahead process when the multi-core processor of a shared-memory type performs a task including a data read-ahead thread for executing data read-ahead and a parallel execution thread for performing an execution process in parallel with the data read-ahead, the system includes a data read-ahead controller which controls an interval between data read-ahead processes in the data read-ahead thread adaptive to a data flow which varies corresponding to an input value of the parallel process in the parallel execution thread. By controlling the interval between the data read-ahead processes, competitions of memory accesses in the multi-core processor are suppressed.
摘要翻译：在共享存储器型的多核处理器中，有效地抑制了来自多个处理器的存储器访问的竞争引起的数据处理能力的恶化。在共享存储器型的多核处理器执行包括用于执行预读数据的数据预读线程的任务的一个用于在数据预读处理中控制对高速缓存存储器的访问的存储器访问控制系统中，并行执行线程，用于与预读数据并行地执行执行处理，该系统包括数据预读控制器，该数据读取控制器控制数据读取线程中的数据预读处理之间的间隔，对应于并行执行线程中并行进程的输入值。通过控制数据预读处理之间的间隔，抑制了多核处理器中存储器访问的竞争。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式