公开(公告)号：US20120005535A1

公开(公告)日：2012-01-05

申请号：US13257121

申请日：2010-02-02

Applicant: Eiichiro Shigehara

Inventor： Eiichiro Shigehara

IPC: G06F11/07

CPC classification number: G06F11/1428 ,  G06F11/0724 ,  G06F11/0751 ,  G06F11/2242

Abstract: A failure diagnostic system (100) for a multicore CPU having installed therein a plurality of CPU cores (11) that is configured to be switched from a SMP mode to an AMP mode includes: load prediction means for predicting a processing load of the multicore CPU; mode switching means for switching at least one of the CPU cores (11) to the AMP mode when the processing load is less than a threshold; and failure diagnostic means for performing a failure diagnosis of the CPU core that has been switched to the AMP mode.

Abstract translation: 一种用于安装有多个CPU核心（11）的多核CPU的故障诊断系统（100），其被配置为从SMP模式切换到AMP模式包括：负载预测装置，用于预测多核CPU的处理负荷 ; 模式切换装置，用于当所述处理负载小于阈值时，将所述CPU核心（11）中的至少一个切换到所述AMP模式; 以及用于执行已经切换到AMP模式的CPU核心的故障诊断的故障诊断装置。

公开(公告)号：US08032791B2

公开(公告)日：2011-10-04

申请号：US12499070

申请日：2009-07-07

Applicant: Sumeet Kochar ,  Barry A. Kritt ,  William B. Schwartz ,  Jeffrey B. Williams

Inventor： Sumeet Kochar ,  Barry A. Kritt ,  William B. Schwartz ,  Jeffrey B. Williams

IPC: G06F11/00

CPC classification number: G06F11/0766 ,  G06F11/0724 ,  G06F11/1417 ,  G06F11/1428 ,  G06F11/2028 ,  G06F11/2041 ,  G06F11/2043

Abstract: Detection of a reset failure in a multinode data processing system is provided by a diagnostic circuit in each of a plurality of the server nodes of the system. Each diagnostic circuit is coupled to a code fetch chain of its corresponding node. At reset, prior to a node processor retrieving startup code from the code fetch chain, the diagnostic circuit provides diagnostic signals to the code fetch chain. A problem in the code fetch chain is detected from a response to the diagnostic signals. When a problem is detected, a node failure status for the problem node may be signaled to the other nodes. The multinode system may be configured in response to signaled node failure status, such as by dropping failed nodes and replacing a failed primary node with a secondary node if necessary.

Abstract translation: 多系统数据处理系统中的复位故障的检测由系统的多个服务器节点中的每一个中的诊断电路提供。 每个诊断电路耦合到其相应节点的代码获取链。 在复位时，在节点处理器从代码获取链中检索启动代码之前，诊断电路向代码获取链提供诊断信号。 从对诊断信号的响应中检测到代码获取链中的问题。 当检测到问题时，问题节点的节点故障状态可以发信号通知给其他节点。 多节点系统可以被配置为响应于信号通知的节点故障状态，例如通过丢弃故障节点并且如果需要则用辅助节点替换故障主节点。

公开(公告)号：US20110154097A1

公开(公告)日：2011-06-23

申请号：US12641072

申请日：2009-12-17

Applicant: Jeffrey A. BARLOW ,  Howard Calkin ,  Andrew C. Walton ,  Anurupa Rajkumari

Inventor： Jeffrey A. BARLOW ,  Howard Calkin ,  Andrew C. Walton ,  Anurupa Rajkumari

IPC: G06F11/20 ,  G06F11/07

CPC classification number: G06F11/079 ,  G06F11/0727 ,  G06F11/0748 ,  G06F11/1428 ,  G06F11/202

Abstract: A system and method for fault management in a computer-based system are disclosed herein. A system includes a plurality of field replaceable units (“FRUs”) and fault management logic. The fault management logic is configured to collect error information from a plurality of components of the system. The logic stores, for each component identified as a possible cause of a detected fault, a record assigning one of two different component failure probability indications. The logic identifies a single of the plurality of FRUs that has failed based on the stored probability indications.

Abstract translation: 本文公开了一种用于基于计算机的系统中的故障管理的系统和方法。 系统包括多个现场可更换单元（“FRU”）和故障管理逻辑。 故障管理逻辑被配置为从系统的多个部件收集错误信息。 对于被识别为检测到的故障的可能原因的每个组件，逻辑存储器分配两个不同组件故障概率指示之一的记录。 该逻辑基于所存储的概率指示来识别已经发生故障的多个FRU中的单个。

公开(公告)号：US20100100886A1

公开(公告)日：2010-04-22

申请号：US12529367

申请日：2008-02-05

Applicant: Masamichi Takagi ,  Masayuki Mizuno ,  Hiroaki Inoue

Inventor： Masamichi Takagi ,  Masayuki Mizuno ,  Hiroaki Inoue

IPC: G06F9/50

CPC classification number: G06F9/5027 ,  G06F11/1428 ,  G06F11/3419 ,  G06F11/3433 ,  G06F2209/483 ,  G06F2209/501

Abstract: Even if a multiprocessor includes an uneven performance core, an inoperative core or a core that does not satisfy such a performance as designed but if the contrivance of task allocation can satisfy the requirement of an application to be executed, the multiple processors are shipped. In a task group allocation method for allocating, to a processor having a plurality of cores, task groups included in an application for the processor to execute, a calculation section measures performances and disposition patterns of the cores, generates a restricting condition associating the measured performances and disposition patterns of the cores with information indicating whether the application can be executed, and, with reference to the restricting condition, reallocates to the cores, the task groups that have previously been allocated to the cores.

Abstract translation: 即使多处理器包括不均衡的性能核心，不能满足设计性能的不工作核心或核心，但如果任务分配的设计能够满足执行应用程序的要求，则会发送多个处理器。 在用于将具有多个核心的处理器分配给应用程序中的处理器执行的任务组的任务组分配方法中，计算部分测量核心的性能和配置模式，产生将测量性能 以及具有指示是否可以执行应用的信息的核心的配置模式，并且参考限制条件将已经分配给核心的任务组重新分配给核心。