公开(公告)号：US20150205676A1

公开(公告)日：2015-07-23

申请号：US14675035

申请日：2015-03-31

Applicant: Huawei Technologies Co., Ltd.

Inventor： Genqiang Zheng ,  Dixuan Zhang

IPC: G06F11/14 ,  G06F9/44

CPC classification number: G06F11/1438 ,  G06F9/4406 ,  G06F11/0721 ,  G06F11/0793 ,  G06F11/1417 ,  G06F11/1441

Abstract: A control method implemented using a server and a server control device. When a server starts abnormally due to an exception of a master central processing unit (CPU), a platform controller hub (PCH) connected to the master CPU, or a Flash that is connected to the PCH connected to the master CPU, the master CPU is reconfigured. It is implemented that, when a startup exception occurs on a server that supports hard partitioning, it can be still ensured that the server normally starts and runs, thereby improving reliability and stability of the server. Therefore, Reliability, Availability and Serviceability (RAS) of the entire server is improved. In addition, a problem that a system fails to run due to an exception of a Flash is solved, and an original dual Basic Input/Output Systems (BIOS) design manner may no longer be used, thereby reducing Flash chip costs of the server.

Abstract translation: 使用服务器和服务器控制设备实现的控制方法。 由于主中央处理单元（CPU）的异常，服务器异常启动时，连接到主CPU的平台控制器集线器（PCH）或连接到连接到主CPU的PCH的闪存，主CPU 被重新配置。 实现了当支持硬分区的服务器发生启动异常时，仍然可以确保服务器正常启动和运行，从而提高服务器的可靠性和稳定性。 因此，提高了整个服务器的可靠性，可用性和可维护性（RAS）。 另外，解决了由于Flash的异常而导致系统无法运行的问题，原来的双基本输入/输出系统（BIOS）设计方式可能不再使用，从而降低了服务器的闪存芯片成本。

公开(公告)号：US10241868B2

公开(公告)日：2019-03-26

申请号：US14675035

申请日：2015-03-31

Applicant: Huawei Technologies Co., Ltd.

Inventor： Genqiang Zheng ,  Dixuan Zhang

IPC: G06F11/00 ,  G06F11/14 ,  G06F9/4401 ,  G06F11/07

Abstract: A control method implemented using a server and a server control device. When a server starts abnormally due to an exception of a master central processing unit (CPU), a platform controller hub (PCH) connected to the master CPU, or a Flash that is connected to the PCH connected to the master CPU, the master CPU is reconfigured. It is implemented that, when a startup exception occurs on a server that supports hard partitioning, it can be still ensured that the server normally starts and runs, thereby improving reliability and stability of the server. Therefore, Reliability, Availability and Serviceability (RAS) of the entire server is improved. In addition, a problem that a system fails to run due to an exception of a Flash is solved, and an original dual Basic Input/Output Systems (BIOS) design manner may no longer be used, thereby reducing Flash chip costs of the server.