(19)
(11)EP 3 306 476 B1

(12)EUROPEAN PATENT SPECIFICATION

(45)Mention of the grant of the patent:
25.03.2020 Bulletin 2020/13

(21)Application number: 16883210.3

(22)Date of filing:  12.09.2016
(51)International Patent Classification (IPC): 
G06F 11/20(2006.01)
G06F 11/14(2006.01)
(86)International application number:
PCT/CN2016/098741
(87)International publication number:
WO 2017/118080 (13.07.2017 Gazette  2017/28)

(54)

METHOD AND APPARATUS FOR HOT CPU REMOVAL AND HOT CPU ADDING DURING OPERATION

VERFAHREN UND VORRICHTUNG ZUM ENTFERNEN UND HINZUFÜGEN EINER ZENTRALEINHEIT (CPU) IM LAUFENDEN BETRIEB

PROCÉDÉ ET DISPOSITIF DE SUPPRESSION ET D'AJOUT A CHAUD D'UNE UNITÉ CENTRALE DE TRAITEMENT (CPU)


(84)Designated Contracting States:
AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

(30)Priority: 08.01.2016 CN 201610016926

(43)Date of publication of application:
11.04.2018 Bulletin 2018/15

(73)Proprietor: Huawei Technologies Co., Ltd.
Longgang District Shenzhen, Guangdong 518129 (CN)

(72)Inventors:
  • ZHANG, Fei
    Shenzhen Guangdong 518129 (CN)
  • LIAO, Defu
    Shenzhen Guangdong 518129 (CN)
  • MA, Zhangping
    Shenzhen Guangdong 518129 (CN)

(74)Representative: Pfenning, Meinig & Partner mbB 
Patent- und Rechtsanwälte Theresienhöhe 11a
80339 München
80339 München (DE)


(56)References cited: : 
CN-A- 1 491 386
CN-A- 103 425 545
US-A1- 2004 215 865
US-A1- 2007 180 288
CN-A- 101 216 793
CN-A- 105 700 975
US-A1- 2006 274 372
US-B1- 6 282 596
  
      
    Note: Within nine months from the publication of the mention of the grant of the European patent, any person may give notice to the European Patent Office of opposition to the European patent granted. Notice of opposition shall be filed in a written reasoned statement. It shall not be deemed to have been filed until the opposition fee has been paid. (Art. 99(1) European Patent Convention).


    Description

    TECHNICAL FIELD



    [0001] The present invention relates to multi-CPU interconnection technologies, and in particular, to a central processing unit (CPU) hot-remove method and apparatus, and a CPU hot-add method and apparatus.

    BACKGROUND



    [0002] With rapid development of the Internet Technology (IT), amounts of data in various IT systems increase. For example, some servers are currently applied to key enterprise services, and because these services are crucial to an enterprise application, data and information processed by the servers are core business data and information of a user, amounts of which are usually massive. For three major types of applications: on-line transaction, business analysis, and a database that are currently most common in some key service fields, even if the three major types of applications are applied to an ordinary enterprise, an amount of data processed in the three major types of applications may be astonishing, not to mention that data at a terabyte (TB) or petabyte (PB) scale needs to be processed when the three major types of applications run in industries such as banking, telecommunications, and securities. Data at such a scale and relations to production, operation, and decision-making efficiency of a business user certainly require a bearing platform to have an extraordinary high-performance processing capability. In addition, with the rise of a large-scale memory database application such as a HANA (High-Performance Analytic Appliance), which is analysis software, an extremely high requirement is posed to a memory capacity of a system with a single server. Therefore, more CPUs and more memories need to be integrated into the single server, so as to meet requirements of high performance and a large capacity that are required for service running.

    [0003] Therefore, a multi-CPU interconnection technology is developed, that is, multiple CPUs are interconnected by using high-speed interconnection channels (for example, a Quick Path Interconnect (QPI)) between the CPUs, so that the multiple physical CPUs can interconnect by using these high-speed interconnection channels and form a resource-sharing server system. However, some extra risks are caused while the multiple CPUs interconnect to improve processing performance of a single server, because the entire system may crash if any CPU in the multi-CPU interconnection system is faulty. If the CPU fault needs to be fixed, the entire server system needs to be powered off, then the CPU is changed, and the operation of powering off the entire server system and changing the CPU certainly causes system service interruption, thereby severely affecting continuous serving time of the system.

    [0004] According to US 2007/180288 A1 in a parallel computing system having a plurality of computing node groups including at least one spare computing node group, a plurality of managing nodes for allocating jobs to the computing node groups and an information management server having respective computing node group status information are associated with the computing node groups, and the respective managing nodes update respective in-use computing node group status information by accessing the information management server. Furthermore, when the managing node detects an occurrence of a failure, the managing node having used then the computing node group disabled due to the failure identifies a spare computing node group by accessing the computing node group status information in the information management server.; Then, the managing node having used then the disabled computing node group obtains the computing node group information of the identified spare computing node group. Furthermore, since the managing node having used then the disabled computing node group can continue processing by switching the disabled computing node group to the identified spare computing node group as a computing node group to be used, on the basis of the computing node group information of the identified spare computing node group, the redundancy in the parallel computing system can be secured.

    [0005] US 6,282,596 B1 describes a method and system for hot-plugging a processor subsystem to a system bus of a data processing system while said data processing system is active, where the system bus comprises multiple positions for supporting hot-pluggable processor subsystems, where each processor subsystem includes a processor and associated voltage regulator module. Power is applied to a processor subsystem in response to an indication that the processor subsystem has been added to the system bus of the data processing system. Initialization routines are performed on the processor within the processor subsystem through a controller which transmits initialization data to the processor independent of the system bus, such that additional processor subsystems are integrated into the data processing system with minimal effect on any existing processors operating on the system bus.

    [0006] CN 103 425 545 A provides a system fault tolerance method for a multiprocessor server. The system fault tolerance method for the multiprocessor server has the following specific steps: a system detects a fault of the processor and reports the fault of the processor to a monitoring and managing unit; the monitoring and managing unit analyzes and judges the fault; after the fault achieves a certain fault level, an interrupt request is sent to the system; and after the system receives the interruption, configuration degradation is carried out according to a fault-tolerance strategy which is formulated in advance. The above steps aim at the multiprocessor server. Compared with the prior art, the system fault tolerance method for the multiprocessor server has the characteristic of improving the system reliability by lowering the configuration as the cost, and has the advantages of strong practicality and easiness in popularizing.

    [0007] In CN 1 491 386 A first and second classes of errors are determined. An error that has led to a reboot of a first node is analyzed and then classified into one of the classes of errors. A cluster is automatically initiated with first and second nodes if the error has been classified into the first class of errors. The cluster for the second class of errors is not initiated automatically.

    SUMMARY



    [0008] Embodiments of the present invention, which is defined in detail in the appended independent claims 1, 3, 5 and 7, provide a central processing unit CPU hot-remove method and apparatus, and a central processing unit CPU hot-add method and apparatus, so that a CPU can be changed without powering-off, and a system can work normally, thereby improving user experience.

    [0009] In an aspect, an embodiment of this application provides a central processing unit CPU hot-remove method. The method is applicable to a server that has a non-full mesh first CPU topology, the server includes a controller, the currently running first CPU topology includes multiple CPUs, and the method may include: determining, by the controller, a first CPU from the multiple CPUs, where the first CPU is a CPU that is faulty or that needs to be removed according to first indication information, and the first indication information is from the first CPU topology or a user interface; determining, by the controller, at least one second CPU from the multiple CPUs, where the at least one second CPU and the first CPU meet a preset condition; and sending, by the controller, second indication information to the first CPU topology, where after the first CPU topology receives the second indication information, the first CPU and the at least one second CPU are removed, so as to obtain a second CPU topology and run the second CPU topology. In this embodiment of the present invention, in-service CPU removal can be achieved, and during a CPU removing process and after the CPU is removed, a system can work normally, thereby improving user experience.

    [0010] In a possible design, multiple CPUs in the first CPU topology may be connected by using an intermediate node, and the intermediate node includes a CPU and/or an external node controller XNC. In this embodiment of the present invention, a CPU in the CPU topology connected by using the intermediate node can be removed online, and the system can work normally, thereby improving user experience.

    [0011] In a possible design, the determining, by the controller, at least one second CPU from the multiple CPUs, where the at least one second CPU and the first CPU meet a preset condition may include: Each CPU in the server may have at least one standby CPU, and the controller may determine at least one standby second CPU of the first CPU. The at least one standby second CPU is in the first CPU topology. According to the present invention, during CPU removing, both the CPU that needs to be removed and the standby CPU may be removed, so that the topology obtained after the CPUs are removed is still stable. The CPUs are removed in a case in which normal system operating is ensured, thereby improving user experience.

    [0012] In a possible design, after the first CPU topology receives the second indication information, resources are reclaimed from the first CPU and the at least one second CPU, and the first CPU and the at least one second CPU are disconnected from a CPU in the second CPU topology; and settings of CPUs in the second CPU topology may be further adjusted, so that remaining CPUs can work in a form of a stable second CPU topology after the first CPU and the at least one second CPU are removed. In this embodiment of the present invention, the CPU topology obtained after the CPUs are removed can work normally, thereby improving user experience.

    [0013] In another aspect, an embodiment of this application provides a central processing unit CPU hot-add method. The method is applicable to a server that has a non-full mesh first CPU topology, the server includes a controller, and the method may include: determining, by the controller, first indication information, where the first indication information is used to instruct to add a first CPU, and the first CPU is not in the currently running first CPU topology; determining, by the controller, whether at least one second CPU is installed, where the at least one second CPU and the first CPU meet a preset condition; and if the at least one second CPU is installed, sending, by the controller, second indication information to the first CPU topology. After the first CPU topology receives the second indication information, the first CPU and the at least one second CPU are added, so as to obtain a second CPU topology and run the second CPU topology. In this embodiment of the present invention, a CPU can be added online, and during an adding process, a system can work normally, thereby improving user experience.

    [0014] In a possible design, the first indication information may be received by using a user interface, and the indication information may carry an identifier of the CPU that needs to be added; or after the first CPU is installed, a sensor triggers a particular instruction, and the controller obtains an identifier of the first CPU according to the instruction. In this embodiment of the present invention, CPU adding may be triggered by using the particular instruction and the user interface, and the system can work normally, thereby improving user experience.

    [0015] In a possible design, the determining, by the controller, whether at least one second CPU is installed, where the at least one second CPU and the first CPU meet a preset condition includes: determining, by the controller, whether at least one standby CPU of the first CPU is installed. In this embodiment of the present invention, both the CPU and the standby CPU can be installed, and the CPU topology is expanded in a case in which an operating system can work normally, thereby improving user experience.

    [0016] In another aspect, an embodiment of this application provides a central processing unit CPU hot-remove apparatus, where the apparatus is applicable to a server that has a non-full mesh first CPU topology, the currently running first CPU topology includes multiple CPUs, and the apparatus includes: a processing unit, configured to determine a first CPU from the multiple CPUs, where the first CPU is a CPU that is faulty or that needs to be removed according to first indication information, and the first indication information is from the first CPU topology or a user interface; where the processing unit is further configured to determine at least one second CPU from the multiple CPUs, where the at least one second CPU and the first CPU meet a preset condition; and a sending unit, configured to send second indication information to the first CPU topology, where the second indication information is used to instruct to remove the first CPU and the at least one second CPU, so as to obtain a second CPU topology and run the second CPU topology.

    [0017] In a possible design, the processing unit is further configured to: determine a location of the first CPU in the first CPU topology, and a second CPU that is in the first CPU topology and that is in at least one location symmetrical to that of the first CPU, or any second CPU that is in a CPU in at least one location symmetrical to that of the first CPU and that is directly connected to the first CPU.

    [0018] In a possible design, each CPU has multiple ports, the multiple CPUs are connected by using the ports, and the processing unit is further configured to: determine at least one second CPU interconnected with the first CPU by using ports of a same port number.

    [0019] In a possible design, the first CPU topology includes multiple CPU groups, information about the multiple CPU groups is pre-stored in the server, and the processing unit is further configured to: determine at least one second CPU belonging to a same CPU group as the first CPU.

    [0020] In a possible design, that the second indication information is used to instruct to remove the first CPU and the at least one second CPU includes: the second indication information is used to instruct to reclaim resources from the first CPU and the at least one second CPU, and disconnect the first CPU and the at least one second CPU from a CPU in the second CPU topology.

    [0021] In still another aspect, an embodiment of this application provides a central processing unit CPU hot-add apparatus, where the apparatus is applicable to a server that has a non-full mesh first CPU topology, and the apparatus includes: a processing unit, configured to determine first indication information, where the first indication information is used to instruct to add a first CPU, and the first CPU is not in the currently running first CPU topology; where the processing unit is further configured to determine whether at least one second CPU is installed, where the at least one second CPU and the first CPU meet a preset condition; and a sending unit, configured to: when the at least one second CPU is installed, send second indication information to the first CPU topology, where the second indication information is used to instruct to add the first CPU and the at least one second CPU, so as to obtain a second CPU topology and run the second CPU topology.

    [0022] In a possible design, the apparatus further includes: a first receiving unit, configured to receive first indication information by using a user interface, where the first indication information includes an identifier of the first CPU; or a second receiving unit, configured to receive, by using a sensor, second indication information triggered when the first CPU is installed; where the processing unit is further configured to determine the installed first CPU according to the second indication information.

    [0023] In a possible design, the processing unit is further configured to: determine whether a second CPU that is in the second CPU topology and that is in at least one location symmetrical to that of the first CPU is installed.

    [0024] In a possible design, the second CPU topology includes multiple CPU groups, information about the multiple CPU groups is pre-stored in the server, and the processing unit is further configured to: determine whether at least one second CPU belonging to a same CPU group as the first CPU is installed.

    [0025] In a possible design, that the second indication information is used to instruct to add the first CPU and the at least one second CPU includes: the second indication information is used to instruct to allocate resources to the first CPU and the at least one second CPU, and establish connections from the first CPU and the at least one second CPU to a CPU in the first CPU topology, so as to obtain the second CPU topology and run the second CPU topology. In the following, the first and second CPUs and CPU topologies of the hot-add method and apparatus are called third and fourth CPUs and CPU topologies to prevent confusion between the first and second CPUs and CPU topologies of the hot-remove method and apparatus and the first and second CPUs and CPU topologies of the hot-add method and apparatus.

    [0026] In still another aspect, an embodiment of this application provides a server that has CPU topology structure, and the server includes a non-full mesh first CPU topology, a controller, and a memory. The memory is configured to store an instruction for the foregoing first aspect, and the controller and the first CPU topology are configured to execute the instruction.

    [0027] In still another aspect, an embodiment of this application provides a server that has a CPU topology structure, and the server includes a non-full mesh third CPU topology, a controller, and a memory. The memory is configured to store an instruction for the foregoing second aspect, and the controller and the third CPU topology are configured to execute the instruction.

    [0028] In still another aspect, an embodiment of this application provides a server that has a CPU topology structure, the structure includes slots, and an independently pluggable CPU is installed in the slot. The slots are connected by using an interconnection channel. The multiple CPUs installed in the slots work as a first CPU topology structure. The server further includes a controller, and the controller is configured to perform the steps in the foregoing first aspect.

    [0029] In still another aspect, an embodiment of this application provides a multipath server that has a CPU topology structure, the structure includes slots, and an independently pluggable CPU is installed in the slot. The slots are connected by using an interconnection channel. The multiple CPUs installed in the slots work as a third CPU topology structure. The server further includes a controller, and the controller is configured to perform the steps in the foregoing second aspect.

    [0030] In still another aspect, an embodiment of the present invention provides a computer storage medium. The computer storage medium is configured to store a computer software instruction used in the foregoing first aspect, and includes a program designed for executing the foregoing aspect.

    [0031] In still another aspect, an embodiment of the present invention provides a computer storage medium. The computer storage medium is configured to store a computer software instruction used in the foregoing second aspect, and includes a program designed for executing the foregoing aspect.

    [0032] According to the CPU hot-remove method and apparatus, and the CPU hot-add method and apparatus that are provided in the embodiments of the present invention, the CPU can be added or removed online, the topology obtained after removing or adding is still a stable topology, and normal system operating is not affected, thereby improving user experience.

    BRIEF DESCRIPTION OF DRAWINGS



    [0033] 

    FIG. 1 is a schematic diagram of a CPU topology structure;

    FIG. 2 is a schematic diagram of another CPU topology structure;

    FIG. 3 is a schematic diagram of a CPU removing process according to an embodiment of the present invention;

    FIG. 4 is a schematic diagram of still another CPU topology structure;

    FIG. 5 is a schematic diagram of yet another CPU topology structure;

    FIG. 6 is a schematic diagram of yet another CPU topology structure;

    FIG. 7 is a schematic diagram of a central processing unit CPU hot-remove method according to an embodiment of the present invention;

    FIG. 8 is a schematic diagram of yet another CPU topology structure;

    FIG. 9 is a schematic diagram of a central processing unit CPU hot-add method according to an embodiment of the present invention;

    FIG. 10 is a schematic structural diagram of a central processing unit CPU hot-remove apparatus according to an embodiment of the present invention;

    FIG. 11 is a schematic structural diagram of a central processing unit CPU hot-add apparatus according to an embodiment of the present invention;

    FIG. 12 is a schematic structural diagram of a server that has a CPU topology according to an embodiment of the present invention; and

    FIG. 13 is a schematic structural diagram of another server that has a CPU topology according to an embodiment of the present invention.


    DESCRIPTION OF EMBODIMENTS



    [0034] To provide thorough understanding of the present invention, the following further describes the embodiments of the present invention in detail with reference to the accompanying drawings. The foregoing embodiments are not intended to limit the protection scope of the present invention.

    [0035] FIG. 1 is a schematic diagram of a CPU topology structure. As shown in FIG. 1, the CPU topology structure may use an Intel processor (Intel Xeon Processor), and the Intel processor includes eight CPUs. The CPUs are connected by using high-speed interconnection channels, and FIG. 1 shows a stable topology structure.

    [0036] In a CPU topology running process, when one CPU is faulty, generally not only the CPU cannot process data, but also all channels connected to the CPU may be faulty. For example, when a CPU 101 shown in FIG. 1 is faulty, a connection between the CPU 101 and a CPU 102, a connection between the CPU 101 and a CPU 103, and a connection between the CPU 101 and a CPU 104 are all faulty. As shown in FIG. 2, FIG. 2 is a schematic diagram of remaining connections that exist when the CPU 101 is faulty. However, a manner in which the seven CPUs are connected shown in FIG. 2 indicates an unstable topology structure. During running, a system fault or crash may be caused.

    [0037] An inventor of this application realizes the problem, and finds, by analyzing, that as shown in FIG. 3, when the CPU 101 needs to be removed, the CPU 103 corresponding to the CPU 101 may be also removed, and in this way, a stable topology structure having six CPUs can be obtained.

    [0038] Therefore, a group of CPUs to which the CPU belongs may be removed to obtain a stable topology structure having fewer than eight CPUs. As shown in FIG. 4, a topology structure having six CPUs may be obtained, or as shown in FIG. 5, a topology structure having four CPUs may be obtained, and so on. The structure in FIG. 4 may be obtained by removing two CPUs in the structure in FIG. 1, and the structure in FIG. 5 may be obtained by removing four CPUs in the structure in FIG. 1. That is, a stable topology structure may be obtained by removing a group of CPUs in a CPU topology structure, and accordingly, a stable topology structure may be obtained by adding a group of CPUs to a CPU topology structure.

    [0039] FIG. 6 is a schematic diagram of a CPU topology structure. As shown in FIG. 6, the CPU topology structure includes eight CPUs, and the CPUs are connected by using high-speed interconnection channels or external node controller (XNC)s. FIG. 6 provides two connection manners using the XNC, but the foregoing problem exists in either connection manner, that is, when one CPU is faulty, connections between remaining seven CPUs are an unstable topology structure. However, whichever CPU is faulty, a CPU corresponding to the faulty CPU can be found, and after the two CPUs are removed, a stable topology structure having six CPUs is obtained.

    [0040] It should be learned that, the stable topology structure having eight CPUs is merely used as an example, and a stable topology structure having another quantity of CPUs also has this feature. For clarity of description, the common stable topology structure having eight CPUs is used.

    [0041] FIG. 7 is a schematic diagram of a central processing unit CPU hot-remove method according to an embodiment of the present invention. As shown in FIG. 3, the method may be performed for a server that has a non-full mesh first CPU topology. An instruction for the following specific steps may be executed by a particular CPU in the first CPU topology, or may be executed by another CPU or a controller different from that in the first CPU topology, and the instruction required for executing the following steps may be stored in a memory. The CPU topology of the server includes multiple CPUs, and the method may include the following steps.

    [0042] S710. Determine a first CPU from the multiple CPUs, where the first CPU is a CPU that is faulty or that needs to be removed according to first indication information, and the first indication information is from the first CPU topology or a user interface.

    [0043] The server may run a service system and a control system, and the service system may perform detection, and determine a CPU that is at risk or that is faulty. The service system is a system that runs on the first CPU topology and that mainly processes a service task, and the control system may be a system that runs on a particular CPU or a controller in the CPU topology and that is mainly used to control the CPU topology. Alternatively, in a first CPU topology running process, when there are a relatively small quantity of tasks and some CPUs need to be stopped from working to save resources, the first CPU topology determines the CPUs that need to be stopped from working. The first CPU topology sends the first indication information to the controller, so as to notify the controller of an identifier of a CPU that needs to be removed. A CPU whose durability or other performance is relatively poor may be removed according to performance of the CPUs.

    [0044] Alternatively, the controller may receive the first indication information by using the user interface. For example, when a CPU needs to be changed, a user may input, by using the user interface, an identifier of the CPU that needs to be changed.

    [0045] The controller may further detect the CPUs in the first topology to determine the faulty CPU, for example, may detect whether the CPU can be normally powered on. In addition, identifiers of the CPUs may be used to distinguish different CPUs, and the identifier of the CPU may be information that can identify the CPU, for example, a socket ID.

    [0046] It should be noted that, for clarity of description, another CPU or a controller different from that in the foregoing first CPU topology is collectively referred to as the controller.

    [0047] S720. Determine at least one second CPU from the multiple CPUs, where the at least one second CPU and the first CPU meet a preset condition.

    [0048] When the at least one second CPU is determined from the multiple CPUs, where the at least one second CPU and the first CPU meet the preset condition, there may be the following implementations:

    Manner 1: CPUs in a same topology structure may have the same type. A CPU module generally has multiple ports, and the ports on the CPU may have different port numbers. However, different CPUs of the same type have same port numbers, and CPUs interconnected by using ports of a same port number may be determined as a CPU group. When determining the at least one second CPU from the multiple CPUs, where the at least one second CPU and the first CPU meet the preset condition, the controller may determine at least one second CPU interconnected with the first CPU by using ports of a same port number. For example, as shown in FIG. 8, the topology structure indicates a topology having eight CPUs, where X (X=0, 1, ..., and 7) in SX represents a socket ID. In addition, 0, 1, and 2 at two ends of each connected line actually represent QPI port numbers. It may be learned from FIG. 8 that, CPU groups including CPUs connected by using ports of a same port number are respectively: S0 and S2, S1 and S3, S4 and S6, and S5 and S7. S0 and S2 are connected by using ports 2, S1 and S3 are connected by using ports 2, S4 and S6 are connected by using ports 2, and S5 and S7 are connected by using ports 2, and each pair of those form a CPU group. When S5 is faulty, a CPU, that is, S7, connected to the port 2 is found, both S5 and S7 are removed, and the remaining CPUs may form a stable topology structure. It should be noted that, the CPUs are grouped according to a stable topology structure rule.

    Manner 2: The controller determines a location of the first CPU in the first CPU topology, and a second CPU that is in the first CPU topology and that is in at least one location symmetrical to that of the first CPU, or any second CPU that is in at least one location symmetrical to that of the first CPU and that is directly connected to the first CPU. Symmetry may be central symmetry or axial symmetry. For example, in the topology in FIG. 3, there are three CPUs in locations symmetrical to that of the CPU 101, two CPUs of the three CPUs are in locations axially symmetrical to that of the CPU 101, and one CPU of the three CPUs is in a location centrally symmetrical to that of the CPU 101. All the three CPUs may be removed, or any CPU directly connected to the CPU 101 may be removed only.

    Manner 3: Each CPU in the server may have at least one standby CPU, and the controller may determine at least one standby second CPU of the first CPU, for example, may group the CPUs in the first CPU topology, and pre-store information about a CPU group in the server; and the controller may determine at least one second CPU belonging to a same CPU group as the first CPU. For another example, each two of CPUs in the topology shown in FIG. 6 may be grouped into a group, so as to form four groups; and identifiers of corresponding CPUs in these CPU groups may be stored, and when one CPU that needs to be removed is determined, the other CPU whose identifier is correspondingly stored is found, and both the CPUs are removed.



    [0049] The service system in the server may determine the at least one second CPU, where the at least one second CPU and the first CPU meet the preset condition. Optionally, the service system in the server sends, to the control system (for example, an operating system (OS), a basic input/output system (BIOS), a baseboard management controller (BMC), or another software), an identifier of a CPU that needs to be removed, and the control system determines a second CPU topology that does not include the first CPU, and sends, to the service system, identifiers of CPUs that need to be removed. The service system removes the corresponding CPUs, so as to obtain the second CPU topology.

    [0050] S730. Send second indication information to the first CPU topology, where the second indication information is used to instruct to remove the first CPU and the at least one second CPU, so as to obtain the second CPU topology and run the second CPU topology.

    [0051] After the CPUs are removed, the server needs to work based on the second CPU topology, for example, may run the service system on the second CPU topology.

    [0052] It should be noted that removing a CPU includes: The system reclaims resources allocated to the CPU, for example, releases the resources allocated to the CPU or moves, to another CPU or another CPU topology, the resources allocated to the CPU, for example, moves the resources to the second CPU topology; the system may further delete logical connections between the CPU that needs to be removed and CPUs in the second CPU topology, that is, remaining CPUs that exist after the CPU that needs to be removed is removed; and the system may further reset the CPUs in the second CPU topology, so that the CPUs can work as the second CPU topology. Further, the CPU that needs to be removed may be further powered off. In the foregoing manner, information for the first CPU and the at least one second CPU does not exist in the CPUs in the second CPU topology, that is, when the system is running, a task that needs to be performed by the first CPU or the at least one second CPU does not exist, and channels to the first CPU and the at least one second CPU are disconnected. Therefore, the second CPU topology can stably run.

    [0053] In a specific implementation of this embodiment of the present invention, CPUs in the CPU topology may be connected by using an intermediate node, and the intermediate node may be a CPU and/or an external node controller XNC, for example, the topology structure shown in FIG. 1 or FIG. 6.

    [0054] In addition, the CPU topology in this embodiment of the present invention may include an even quantity (for example, 8 or 6) of CPUs, and accordingly, a CPU topology obtained after removing still has an even quantity of CPUs.

    [0055] It should be learned that, both the first CPU topology and the second CPU topology are stable topology structures.

    [0056] In this embodiment of the present invention, when the CPU is faulty or needs to be removed, the CPU can be removed in a case in which normal system operating is not affected and the CPU topology obtained after removing remains stable, thereby improving user experience.

    [0057] For a server that has a non-full mesh CPU topology, the server not only needs to have a continuous serving capability, but also needs to have a flexible expansion capability. The expansion may be: when hardware resources are insufficient or performance is not good enough, increasing hardware resources of the server to expand system resources, so as to improve the server performance. This process is referred to as expansion. The following provides a CPU topology expansion method.

    [0058] FIG. 9 is a schematic diagram of a central processing unit CPU hot-add method according to an embodiment of the present invention. As shown in FIG. 9, the method may be performed for a multipath server that has a non-full mesh CPU topology. An instruction for the following specific steps may be executed by a particular CPU in the non-full mesh CPU topology, or may be executed by another CPU or a controller different from that in the non-full mesh CPU topology, and the instruction required for executing the following steps may be stored in a memory. The method may include the following steps.

    [0059] S910. Determine first indication information, where the first indication information is used to instruct to add a third CPU, and the third CPU is not in the currently running third CPU topology.

    [0060] After completing installation of the third CPU, the user may input an instruction by using a user interface, and the controller may receive the instruction. The instruction may carry an identifier of the third CPU.

    [0061] Alternatively, after a CPU that needs to be added is installed, a sensor triggers a particular electrical signal, and the controller may receive the signal, and obtain an identifier of the third CPU according to an indication of the electrical signal. The identifier of the CPU may be information that can identify the CPU, for example, a socket ID. For example, different electrical signals may be triggered by using different slots, and a slot in which the CPU is installed may be determined by using the electrical signal. Alternatively, electrical signals triggered by using different slots may be the same, and after receiving the electrical signal, the server may learn that a new CPU is installed, and may determine an identifier of the newly installed CPU by using a service system or a control system.

    [0062] S920. Determine whether at least one fourth CPU is installed, where the at least one fourth CPU and the third CPU meet a preset condition .

    [0063] To determine whether the at least one fourth CPU is installed, where the at least one fourth CPU and the third CPU meet the preset condition, the following specific implementations may be performed.

    [0064] Manner 1: A principle of this manner is the same as a principle of Manner 2 in step S720 shown in FIG. 7, and may be understood by referring to Manner 2 in step S720. A specific step is as follows: The controller determines whether a fourth CPU that is in a fourth CPU topology and that is in at least one location symmetrical to that of the third CPU is installed.

    [0065] Manner 2: A principle of this manner is the same as a principle of Manner 3 in step S720 shown in FIG. 7, and may be understood by referring to Manner 3 in step S720. A specific step is as follows: The processor may determine whether at least one standby CPU of the third CPU is installed. For example, the fourth CPU topology includes multiple CPU groups, information about the multiple CPU groups may be pre-stored in the server, and the controller determines whether at least one fourth CPU belonging to a same CPU group as the third CPU is installed.

    [0066] A group of CPUs needs to be added to the third CPU topology, so as to obtain a stable topology. When hot-add indication information is determined, there may be an identifier of only one CPU, and the service system needs to determine an in-position status of another CPU corresponding to the identifier of the CPU; and when both the CPU and the corresponding CPU are installed, the following step is further performed.

    [0067] S930. If the at least one fourth CPU is installed, where the at least one fourth CPU and the third CPU meet the preset condition, send second indication information to the third CPU topology, where the second indication information is used to instruct to add the third CPU and the at least one fourth CPU, so as to obtain a fourth CPU topology and run the fourth CPU topology.

    [0068] After receiving the second indication information, the third CPU topology allocates resources to the third CPU and the at least one fourth CPU, and establishes connections from the third CPU and the at least one fourth CPU to a CPU in the third CPU topology; and may further adjust settings of CPUs in the third CPU topology, so that the CPUs in the third CPU topology, the third CPU, and the at least one fourth CPU can operate as the fourth CPU topology.

    [0069] It should be learned that, both the third CPU topology and the fourth CPU topology are stable topology structures.

    [0070] In this embodiment of the present invention, a capacity of the CPU topology can be expanded without affecting normal system operating, or with reference to the embodiment shown in FIG. 7 and the embodiment shown in FIG. 9, a CPU can be changed, so that a system runs more stably, and user experience is further improved.

    [0071] The foregoing mainly describes the solutions provided in this embodiment of the present invention from a perspective of a data processing procedure of the multipath server. It may be understood that, to achieve the foregoing functions, the server includes a corresponding hardware structure and/or software module for implementing each function. A person of ordinary skill in the art should be easily aware that, the units and algorithm steps in each example described with reference to the embodiments disclosed in this specification may be implemented in a form of hardware or a combination of hardware and computer software in the present invention. Whether the functions are implemented by hardware or are implemented in a manner in which computer software drives hardware depends on particular applications and design constraint requirements of the technical solutions. A person skilled in the art may use different methods to implement the described functions for each particular application, but it should not be considered that the implementation goes beyond the scope of the present invention.

    [0072] FIG. 10 is a schematic structural diagram of a central processing unit CPU hot-remove apparatus according to an embodiment of the present invention. The apparatus is applicable to a server that has a non-full mesh first CPU topology, the currently running first CPU topology includes multiple CPUs, and the apparatus includes:

    a processing unit 1001, configured to determine a first CPU from the multiple CPUs, where the first CPU is a CPU that is faulty or that needs to be removed according to first indication information, and the first indication information is from the first CPU topology or a user interface; where

    the processing unit 1001 is further configured to determine at least one second CPU from the multiple CPUs, where the at least one second CPU and the first CPU meet a preset condition; and

    a sending unit 1002, configured to send second indication information to the first CPU topology, where the second indication information is used to instruct to remove the first CPU and the at least one second CPU, so as to obtain a second CPU topology and run the second CPU topology.



    [0073] Optionally, the processing unit 1001 is further configured to:

    determine a location of the first CPU in the first CPU topology, and a second CPU that is in the first CPU topology and that is in at least one location symmetrical to that of the first CPU, or any second CPU that is in at least one location symmetrical to that of the first CPU and that is directly connected to the first CPU. Optionally, each CPU has multiple ports, the multiple CPUs are connected by using the ports, and the processing unit 1001 is further configured to:

    determine at least one second CPU interconnected with the first CPU by using ports of a same port number.



    [0074] Optionally, the processing unit 1001 is further configured to: determine at least one standby second CPU of the first CPU.

    [0075] Further, the first CPU topology includes multiple CPU groups, information about the multiple CPU groups is pre-stored in the server, and the processing unit 1001 is further configured to:
    determine at least one second CPU belonging to a same CPU group as the first CPU.

    [0076] Optionally, that the second indication information is used to instruct to remove the first CPU and the at least one second CPU includes:
    the second indication information is used to instruct to reclaim resources from the first CPU and the at least one second CPU, and disconnect the first CPU and the at least one second CPU from a CPU in the second CPU topology.

    [0077] It should be noted that, this embodiment of the present invention is corresponding to the foregoing method embodiment shown in FIG. 7, and reference may be made to each other for understanding. Details are not described.

    [0078] FIG. 11 is a schematic structural diagram of a central processing unit CPU hot-add apparatus according to an embodiment of the present invention. The apparatus is applicable to a server that has a non-full mesh third CPU topology, and the apparatus includes:

    a processing unit 1101, configured to determine first indication information, where the first indication information is used to instruct to add a third CPU, and the third CPU is not in the currently running third CPU topology; where

    the processing unit 1101 is further configured to determine whether at least one fourth CPU is installed, where the at least one fourth CPU and the third CPU meet a preset condition; and

    a sending unit 1102, configured to: when the at least one fourth CPU is installed, send second indication information to the third CPU topology, where the second indication information is used to instruct to add the third CPU and the at least one fourth CPU, so as to obtain a fourth CPU topology and run the fourth CPU topology.



    [0079] Optionally, the apparatus further includes:

    a first receiving unit, configured to receive third indication information by using a user interface, where the third indication information includes an identifier of the third CPU; or

    a second receiving unit, configured to receive, by using a sensor, fourth indication information triggered when the third CPU is installed; where the processing unit 1101 is further configured to determine the installed third CPU according to the fourth indication information.



    [0080] Optionally, the processing unit 1101 is further configured to:

    determine whether a fourth CPU that is in the fourth CPU topology and that is in at least one location symmetrical to that of the third CPU is installed. Optionally, the processing unit 1101 is further configured to:

    determine at least one standby second CPU of the first CPU.



    [0081] Further, the fourth CPU topology includes multiple CPU groups, information about the multiple CPU groups is pre-stored in the server, and the processing unit 1101 is further configured to:
    determine whether at least one fourth CPU belonging to a same CPU group as the third CPU is installed.

    [0082] Optionally, that the second indication information is used to instruct to add the third CPU and the at least one fourth CPU includes: the second indication information is used to instruct to allocate resources to the third CPU and the at least one fourth CPU, and establish connections from the third CPU and the at least one fourth CPU to a CPU in the third CPU topology, so as to obtain the fourth CPU topology and run the fourth CPU topology.

    [0083] It should be noted that, this embodiment of the present invention is corresponding to the foregoing method embodiment shown in FIG. 9, and reference may be made to each other for understanding. Details are not described.

    [0084] FIG. 12 is a schematic structural diagram of a server that has a CPU topology structure according to an embodiment of the present invention. The server may include a CPU topology 1201, an input/output interface 1202, and a memory 1203 and a bus 1204 that are further shown in the figure, and may further include a controller 1205. The CPU topology 1201, the input/output interface 1202, the memory 1203, and the controller 1205 are connected and communicate with each other by using the bus 1204. The memory 1203 is configured to store a program, and the CPU topology 1201 and the controller 1205 execute the program by reading the program stored in the memory, and send and receive data and an instruction for an external device by using the input/output interface 1202.

    [0085] It should be noted that a CPU topology structure of the CPU topology 1201 herein includes slots, and an independently pluggable CPU is installed in the slot. The slots are connected by using an interconnection channel, so as to form a stable topology structure. The multiple CPUs installed in the slots work as a first CPU topology structure.

    [0086] A CPU corresponding to a to-be-removed CPU generally exists in the first CPU topology, and the slots may be used to distinguish the to-be-removed CPU and the corresponding CPU from another CPU. For example, if the to-be-removed and the CPU corresponding to the to-be-removed CPU are considered as a CPU group, slots belonging to a same slot group may be identified by using a same identifier or a same type of identifiers, or a same group of slots may be circled in a same block on a main board, or a same group of slots may be marked by using a same color.

    [0087] The memory 1203 may be one storage apparatus or may be a collective term of multiple storage elements, and is configured to store executable program code in the foregoing steps, or a parameter, data, and the like that are required for running an access network management device. In addition, the memory 1203 may include a random access memory (RAM), or may include a non-volatile memory, such as a magnetic disk memory, a flash.

    [0088] The bus 1204 may be an industry standard architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an extended industry standard architecture (EISA) bus, or the like. The bus 1204 may be classified into an address bus, a data bus, a control bus, or the like. For ease of representation, only one bold line is used in FIG. 12 for representation, but it does not indicate that there is only one bus or one type of bus.

    [0089] FIG. 13 is a schematic structural diagram of another server that has a CPU topology structure according to an embodiment of the present invention. The multipath server may include a CPU topology 1301, an input/output interface 1302, and a memory 1303 and a bus 1304 that are further shown in the figure, and may further include a controller 1305. The CPU topology 1301, the input/output interface 1302, the memory 1303, and the controller 1305 are connected and communicate with each other by using the bus 1304. It should be noted that a CPU topology structure of the CPU topology 1301 herein includes slots, and an independently pluggable CPU is installed in the slot. The slots are connected by using an interconnection channel, so as to form a stable third CPU topology.

    [0090] A CPU corresponding to a to-be-removed CPU generally exists in a fourth CPU topology, and slots may be reserved in the third CPU topology. A to-be-added CPU and a CPU corresponding to the to-be-added CPU may be installed in the reserved slots. The reserved slots may be distinguished from the slots in the fourth CPU topology. For example, if the to-be-added CPU and the CPU corresponding to the to-be-added CPU are considered as a CPU group, slots belonging to a same slot group may be identified by using a same identifier or a same type of identifiers, or a same group of slots may be circled in a same block on a main board, or a same group of slots may be marked by using a same color.

    [0091] The foregoing modules are similar to the modules in FIG. 12, and reference may be made to each other for understanding. Details are not described.

    [0092] In this embodiment of the present invention, CPU hot plug may be implemented without affecting stability of the CPU topology, so that a system can operate normally, thereby improving user experience.

    [0093] A person skilled in the art may be further aware that, in combination with the examples described in the embodiments disclosed in this specification, units and algorithm steps may be implemented by electronic hardware, computer software, or a combination thereof. To clearly describe the interchangeability between the hardware and the software, the foregoing has generally described compositions and steps of each example according to functions. Whether the functions are performed by hardware or software depends on particular applications and design constraint conditions of the technical solutions. A person skilled in the art may use different methods to implement the described functions for each particular application, but it should not be considered that the implementation goes beyond the scope of the present invention.

    [0094] Steps of methods or algorithms described in the embodiments disclosed in this specification may be implemented by hardware, a software module executed by a processor, or a combination thereof. The software module may reside in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

    [0095] In the foregoing specific implementations, the objective, technical solutions, and benefits of the present invention are further described in detail. It should be understood that the foregoing descriptions are merely specific implementations of the present invention, but are not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, or improvement made without departing from the principle of the present invention should fall within the protection scope of the present invention.


    Claims

    1. A central processing unit, CPU, hot-remove method, wherein the method is applicable to a single server that has a non-full mesh first CPU topology, the server comprises a controller, the currently running first CPU topology comprises multiple CPUs, wherein the method comprises:

    determining (710), by the controller, a first CPU from the multiple CPUs, wherein the first CPU is a CPU that is faulty or that needs to be removed according to first indication information, and the first indication information is from the first CPU topology or a user interface;

    determining (720), by the controller, at least one second CPU from the multiple CPUs, wherein the at least one second CPU and the first CPU meet a preset condition; and

    sending (730), by the controller, second indication information to the first CPU topology, wherein the second indication information is used to instruct to remove the first CPU and the at least one second CPU, so as to obtain a second CPU topology and run the second CPU topology;

    wherein the determining (710), by the controller, at least one second CPU from the multiple CPUs, wherein the at least one second CPU and the first CPU meet a preset condition comprises:
    determining, by the controller, at least one standby second CPU of the first CPU, wherein the at least one standby second CPU is the at least one second CPU.


     
    2. The method according to claim 1, wherein that the second indication information is used to instruct to remove the first CPU and the at least one second CPU comprises:
    the second indication information is used to instruct to reclaim resources from the first CPU and the at least one second CPU in the first CPU topology, and disconnect the first CPU and the at least one second CPU from a CPU in the second CPU topology.
     
    3. A central processing unit, CPU, hot-add method, wherein the method is applicable to a single server that has a non-full mesh first CPU topology, the server comprises a controller, wherein the method comprises:

    determining (910), by the controller, first indication information, wherein the first indication information is used to instruct to add a first CPU, and the first CPU is not in the currently running first CPU topology;

    determining (920), by the controller, whether at least one second CPU is installed, wherein the at least one second CPU and the first CPU meet a preset condition; and

    sending (930), by the controller, second indication information to the first CPU topology when the at least one second CPU is installed, wherein the second indication information is used to instruct to add the first CPU and the at least one second CPU, so as to obtain a second CPU topology and run the second CPU topology;

    wherein the determining, by the controller, whether at least one second CPU is installed, where the at least one second CPU and the first CPU meet a preset condition comprises:
    determining, by the controller, whether at least one standby second CPU of the first CPU is installed, wherein the at least one standby second CPU is the at least one second CPU.


     
    4. The method according to claim 3, wherein the determining, by the controller, first indication information comprises:

    receiving, by the controller, first indication information by using a user interface, wherein the first indication information comprises an identifier of the first CPU; or

    receiving, by the controller by using a sensor, second indication information triggered when the first CPU is installed, and determining the installed first CPU according to the second indication information.


     
    5. A central processing unit, CPU, hot-remove apparatus, wherein the apparatus is applicable to a single server that has a non-full mesh first CPU topology, the currently running first CPU topology comprises multiple CPUs, wherein the apparatus comprises:

    a processing unit (1001), configured to determine a first CPU from the multiple CPUs, wherein the first CPU is a CPU that is faulty or that needs to be removed according to first indication information, and the first indication information is from the first CPU topology or a user interface; wherein

    the processing unit (1001) is further configured to determine at least one second CPU from the multiple CPUs, wherein the at least one second CPU and the first CPU meet a preset condition; and

    a sending unit (1002), configured to send second indication information to the first CPU topology, wherein the second indication information is used to instruct to remove the first CPU and the at least one second CPU, so as to obtain a second CPU topology and run the second CPU topology;

    wherein the processing unit is further configured to determine at least one standby second CPU of the first CPU, wherein the at least one standby second CPU is the at least one second CPU.


     
    6. The apparatus according to claim 5, wherein that the second indication information is used to instruct to remove the first CPU and the at least one second CPU comprises:
    the second indication information is used to instruct to reclaim resources from the first CPU and the at least one second CPU in the first CPU topology, and disconnect the first CPU and the at least one second CPU from a CPU in the second CPU topology.
     
    7. A central processing unit, CPU, hot-add apparatus, wherein the apparatus is applicable to a single server that has a non-full mesh first CPU topology, wherein the apparatus comprises:

    a processing unit (1101), configured to determine first indication information, wherein the first indication information is used to instruct to add a first CPU, and the first CPU is not in the currently running first CPU topology; wherein

    the processing unit (1101) is further configured to determine whether at least one second CPU is installed, where the at least one second CPU and the first CPU meet a preset condition; and

    a sending unit (1102), configured to: when the at least one second CPU is installed, send second indication information to the first CPU topology, wherein the second indication information is used to instruct to add the first CPU and the at least one second CPU, so as to obtain a second CPU topology and run the second CPU topology;

    wherein the processing unit is further configured to:
    determine whether at least one standby second CPU of the first CPU is installed, wherein the at least one standby second CPU is the at least one second CPU.


     
    8. The apparatus according to claim 7, further comprising:

    a first receiving unit, configured to receive first indication information by using a user interface, wherein the first indication information comprises an identifier of the first CPU; or

    a second receiving unit, configured to receive, by using a sensor, second indication information triggered when the first CPU is installed; wherein the processing unit is further configured to determine the installed first CPU according to the second indication information.


     


    Ansprüche

    1. Verfahren zum Entfernen einer Zentraleinheit (Central Processing Unit, CPU) im laufenden Betrieb, wobei das Verfahren auf einen einzelnen Server, der eine erste CPU-Topologie mit einer nicht vollständigen Vernetzung aufweist, anwendbar ist, wobei der Server eine Steuerung umfasst, die erste CPU-Topologie, die aktuell ausgeführt wird, mehrere CPUs umfasst, und wobei das Verfahren Folgendes umfasst:

    Bestimmen (710) durch die Steuerung einer ersten CPU aus den mehreren CPUs, wobei es sich bei der ersten CPU um eine CPU handelt, die fehlerhaft ist, oder die gemäß ersten Anzeigeinformationen entfernt werden muss, und wobei die ersten Anzeigeinformationen von der ersten CPU-Topologie oder einer Benutzerschnittstelle stammen;

    Bestimmen (720) durch die Steuerung von mindestens einer zweiten CPU aus den mehreren CPUs, wobei die mindestens eine zweite CPU und die erste CPU eine vorgewählte Bedingung erfüllen; und

    Senden (730) durch die Steuerung von zweiten Anzeigeinformationen an die erste CPU-Topologie, wobei die zweiten Anzeigeinformationen dazu verwendet werden, dazu anzuweisen, die erste CPU und die mindestens eine zweite CPU zu entfernen, um eine zweite CPU-Topologie zu erhalten und die zweite CPU-Topologie auszuführen;

    wobei das Bestimmen (710) durch die Steuerung von mindestens einer zweiten CPU aus den mehreren CPUs, wobei die mindestens eine zweite CPU und die erste CPU eine vorgewählte Bedingung erfüllen, Folgendes umfasst:
    Bestimmen durch die Steuerung von mindestens einer zweiten Standby-CPU der ersten CPU, wobei es sich bei der mindestens einen zweiten Standby-CPU um die mindestens eine zweite CPU handelt.


     
    2. Verfahren nach Anspruch 1, wobei, dass die zweiten Anzeigeinformationen dazu verwendet werden, dazu anzuweisen, die erste CPU und die mindestens eine zweite CPU zu entfernen, Folgendes umfasst:
    dass die zweiten Anzeigeinformationen dazu verwendet werden, dazu anzuweisen, Ressourcen von der ersten CPU und der mindestens einen zweiten CPU in der ersten CPU-Topologie zurückzugewinnen und die erste CPU und die mindestens eine zweite CPU von einer CPU in der zweiten CPU-Topologie zu trennen.
     
    3. Verfahren zum Hinzufügen einer Zentraleinheit (CPU) im laufenden Betrieb, wobei das Verfahren auf einen einzelnen Server, der eine erste CPU-Topologie mit einer nicht vollständigen Vernetzung aufweist, anwendbar ist, wobei der Server eine Steuerung umfasst, wobei das Verfahren Folgendes umfasst:

    Bestimmen (910) von ersten Anzeigeinformationen durch die Steuerung, wobei die ersten Anzeigeinformationen dazu verwendet werden, dazu anzuweisen, eine erste CPU hinzuzufügen, und wobei die erste CPU nicht in der ersten CPU-Topologie, die aktuell ausgeführt wird, vorhanden ist;

    Bestimmen (920) durch die Steuerung, ob mindestens eine zweite CPU installiert wird, wobei die mindestens eine zweite CPU und die erste CPU eine vorgewählte Bedingung erfüllen; und

    Senden (930) von zweiten Anzeigeinformationen an die erste CPU-Topologie durch die Steuerung, wenn die mindestens eine zweite CPU installiert wird, wobei die zweiten Anzeigeinformationen dazu verwendet werden, dazu anzuweisen, die erste CPU und die mindestens eine zweite CPU hinzuzufügen, um eine zweite CPU-Topologie zu erhalten und die zweite CPU-Topologie auszuführen;

    wobei das Bestimmen durch die Steuerung, ob mindestens eine zweite CPU installiert wird, wobei die mindestens eine zweite CPU und die erste CPU eine vorgewählte Bedingung erfüllen, Folgendes umfasst:
    Bestimmen durch die Steuerung, ob mindestens eine zweite Standby-CPU der ersten CPU installiert wird, wobei es sich bei der mindestens einen zweiten Standby-CPU um die mindestens eine zweite CPU handelt.


     
    4. Verfahren nach Anspruch 3, wobei das Bestimmen der ersten Anzeigeinformationen durch die Steuerung Folgendes umfasst:

    Empfangen durch die Steuerung von ersten Anzeigeinformationen unter Verwendung einer Benutzerschnittstelle, wobei die ersten Anzeigeinformationen eine Kennung der ersten CPU umfassen; oder

    Empfangen durch die Steuerung unter Verwendung eines Sensors von zweiten Anzeigeinformationen, die ausgelöst werden, wenn die erste CPU installiert wird, und Bestimmen der installierten ersten CPU gemäß den zweiten Anzeigeinformationen.


     
    5. Vorrichtung zum Entfernen einer Zentraleinheit (CPU) im laufenden Betrieb, wobei die Vorrichtung auf einen einzelnen Server, der eine erste CPU-Topologie mit einer nicht vollständigen Vernetzung aufweist, anwendbar ist, wobei die erste CPU-Topologie, die aktuell ausgeführt wird, mehrere CPUs umfasst, wobei die Vorrichtung Folgendes umfasst:

    eine Verarbeitungseinheit (1001), die dazu konfiguriert ist, eine erste CPU aus den mehreren CPUs zu bestimmen, wobei es sich bei der ersten CPU um eine CPU handelt, die fehlerhaft ist, oder die gemäß ersten Anzeigeinformationen entfernt werden muss, und wobei die ersten Anzeigeinformationen von der ersten CPU-Topologie oder einer Benutzerschnittstelle stammen; wobei

    die Verarbeitungseinheit (1001) ferner dazu konfiguriert ist, mindestens eine zweite CPU aus den mehreren CPUs zu bestimmen, wobei die mindestens eine zweite CPU und die erste CPU eine vorgewählte Bedingung erfüllen; und

    eine Sendeeinheit (1002), die dazu konfiguriert ist, zweite Anzeigeinformationen an die erste CPU-Topologie zu senden, wobei die zweiten Anzeigeinformationen dazu verwendet werden, dazu anzuweisen, die erste CPU und die mindestens eine zweite CPU zu entfernen, um eine zweite CPU-Topologie zu erhalten und die zweite CPU-Topologie auszuführen;

    wobei die Verarbeitungseinheit ferner dazu konfiguriert ist, mindestens eine zweite Standby-CPU der ersten CPU zu bestimmen, wobei es sich bei der mindestens einen zweiten Standby-CPU um die mindestens eine zweite CPU handelt.


     
    6. Vorrichtung nach Anspruch 5, wobei, dass die zweiten Anzeigeinformationen dazu verwendet werden, dazu anzuweisen, die erste CPU und die mindestens eine zweite CPU zu entfernen, Folgendes umfasst:
    dass die zweiten Anzeigeinformationen dazu verwendet werden, dazu anzuweisen, Ressourcen von der ersten CPU und der mindestens einen zweiten CPU in der ersten CPU-Topologie zurückzugewinnen, und die erste CPU und die mindestens eine zweite CPU von einer CPU in der zweiten CPU-Topologie zu trennen.
     
    7. Vorrichtung zum Hinzufügen einer Zentraleinheit (CPU) im laufenden Betrieb, wobei die Vorrichtung auf einen einzelnen Server, der eine erste CPU-Topologie mit einer nicht vollständigen Vernetzung aufweist, anwendbar ist, wobei die Vorrichtung Folgendes umfasst:

    eine Verarbeitungseinheit (1101), die dazu konfiguriert ist, erste Anzeigeinformationen zu bestimmen, wobei die ersten Anzeigeinformationen dazu verwendet werden, dazu anzuweisen, eine erste CPU hinzuzufügen, und wobei die erste CPU nicht in der ersten CPU-Topologie vorhanden ist, die aktuell ausgeführt wird; wobei

    die Verarbeitungseinheit (1101) ferner dazu konfiguriert ist, zu bestimmen, ob mindestens eine zweite CPU installiert wird, wobei die mindestens eine zweite CPU und die erste CPU eine vorgewählte Bedingung erfüllen; und

    eine Sendeeinheit (1102), die zu Folgendem konfiguriert ist: wenn die mindestens eine zweite CPU installiert wird, Senden von zweiten Anzeigeinformationen an die erste CPU-Topologie, wobei die zweiten Anzeigeinformationen dazu verwendet werden, dazu anzuweisen, die erste CPU und die mindestens eine zweite CPU hinzuzufügen, um eine zweite CPU-Topologie zu erhalten und die zweite CPU-Topologie auszuführen;

    wobei die Verarbeitungseinheit ferner zu Folgendem konfiguriert ist:
    Bestimmen, ob mindestens eine zweite Standby-CPU der ersten CPU installiert wird, wobei es sich bei der mindestens einen zweiten Standby-CPU um die mindestens eine zweite CPU handelt.


     
    8. Vorrichtung nach Anspruch 7, ferner umfassend:

    eine erste Empfangseinheit, die dazu konfiguriert ist, erste Anzeigeinformationen unter Verwendung einer Benutzerschnittstelle zu empfangen, wobei die ersten Anzeigeinformationen eine Kennung der ersten CPU umfassen; oder

    eine zweite Empfangseinheit, die dazu konfiguriert ist, unter Verwendung eines Sensors zweite Anzeigeinformationen zu empfangen, die ausgelöst werden, wenn die erste CPU installiert wird; wobei die Verarbeitungseinheit ferner dazu konfiguriert ist, die installierte erste CPU gemäß den zweiten Anzeigeinformationen zu bestimmen.


     


    Revendications

    1. Procédé de retrait à chaud d'une unité centrale de traitement (CPU), dans lequel le procédé est applicable à un serveur unique qui présente une première topologie de CPU à maillage incomplet, le serveur comprend un contrôleur, la première topologie de CPU en cours d'exploitation comprend de multiples CPU, dans lequel le procédé comprend :

    la détermination (710), par le contrôleur, d'une première CPU parmi les multiples CPU, dans lequel la première CPU est une CPU qui est défectueuse ou qui doit être retirée selon des premières informations d'indication, et les premières informations d'indication proviennent de la première topologie de CPU ou d'une interface utilisateur ;

    la détermination (720), par le contrôleur, d'au moins une seconde CPU parmi les multiples CPU, dans lequel l'au moins une seconde CPU et la première CPU remplissent une condition prédéfinie ; et

    l'envoi (730), par le contrôleur, de secondes informations d'indication à la première topologie de CPU, dans lequel les secondes informations d'indication sont utilisées pour demander le retrait de la première CPU et de l'au moins une seconde CPU, de manière à obtenir une seconde topologie de CPU et exploiter la seconde topologie de CPU ;

    dans lequel la détermination (710), par le contrôleur, d'au moins une seconde CPU parmi les multiples CPU, dans lequel l'au moins une seconde CPU et la première CPU remplissent une condition prédéfinie, comprend :
    la détermination, par le contrôleur, d'au moins une seconde CPU de secours de la première CPU, dans lequel l'au moins une seconde CPU de secours est l'au moins une seconde CPU.


     
    2. Procédé selon la revendication 1, dans lequel l'utilisation des secondes informations d'indication pour demander le retrait de la première CPU et de l'au moins une seconde CPU comprend :
    l'utilisation des secondes informations d'indication pour demander la récupération de ressources depuis la première CPU et l'au moins une seconde CPU de la première topologie de CPU, et la déconnexion de la première CPU et de l'au moins une seconde CPU d'une CPU de la seconde topologie de CPU.
     
    3. Procédé d'ajout à chaud d'une unité centrale de traitement (CPU), dans lequel le procédé est applicable à un serveur unique qui présente une première topologie de CPU à maillage incomplet, le serveur comprend un contrôleur, dans lequel le procédé comprend :

    la détermination (910), par le contrôleur, de premières informations d'indication, dans lequel les premières informations d'indication sont utilisées pour demander l'ajout d'une première CPU, et la première CPU n'est pas dans la première topologie de CPU en cours d'exploitation ;

    le fait de déterminer (920), par le contrôleur, si au moins une seconde CPU est installée, dans lequel l'au moins une seconde CPU et la première CPU remplissent une condition prédéfinie ; et

    l'envoi (930), par le contrôleur, de secondes informations d'indication à la première topologie de CPU lorsque l'au moins une seconde CPU est installée, dans lequel les secondes informations d'indication sont utilisées pour demander l'ajout de la première CPU et de l'au moins une seconde CPU, de manière à obtenir une seconde topologie de CPU et exploiter la seconde topologie de CPU ;

    dans lequel le fait de déterminer, par le contrôleur, si au moins une seconde CPU est installée, où l'au moins une seconde CPU et la première CPU remplissent une condition prédéfinie, comprend :
    le fait de déterminer, par le contrôleur, si au moins une seconde CPU de secours de la première CPU est installée, dans lequel l'au moins une seconde CPU de secours est l'au moins une seconde CPU.


     
    4. Procédé selon la revendication 3, dans lequel la détermination, par le contrôleur, de premières informations d'indication comprend :

    la réception, par le contrôleur, de premières informations d'indication au moyen d'une interface utilisateur, dans lequel les premières informations d'indication comprennent un identifiant de la première CPU ; ou

    la réception, par le contrôleur au moyen d'un capteur, de secondes informations d'indication déclenchées lorsque la première CPU est installée, et la détermination de la première CPU installée selon les secondes informations d'indication.


     
    5. Appareil de retrait à chaud d'une unité centrale de traitement (CPU), dans lequel l'appareil est applicable à un serveur unique qui présente une première topologie de CPU à maillage incomplet, la première topologie de CPU en cours d'exploitation comprend de multiples CPU, dans lequel l'appareil comprend :

    une unité de traitement (1001), configurée pour déterminer une première CPU parmi les multiples CPU, dans lequel la première CPU est une CPU qui est défectueuse ou qui doit être retirée selon des premières informations d'indication, et les premières informations d'indication proviennent de la première topologie de CPU ou d'une interface utilisateur ; dans lequel l'unité de traitement (1001) est configurée en outre pour déterminer au moins une seconde CPU parmi les multiples CPU, dans lequel l'au moins une seconde CPU et la première CPU remplissent une condition prédéfinie ; et

    une unité d'envoi (1002), configurée pour envoyer des secondes informations d'indication à la première topologie de CPU, dans lequel les secondes informations d'indication sont utilisées pour demander le retrait de la première CPU et de l'au moins une seconde CPU, de manière à obtenir une seconde topologie de CPU et exploiter la seconde topologie de CPU ;

    dans lequel l'unité de traitement est configurée en outre pour déterminer au moins une seconde CPU de secours de la première CPU, dans lequel l'au moins une seconde CPU de secours est l'au moins une seconde CPU.


     
    6. Appareil selon la revendication 5, dans lequel l'utilisation des secondes informations d'indication pour demander le retrait de la première CPU et de l'au moins une seconde CPU comprend :
    l'utilisation des secondes informations d'indication pour demander la récupération de ressources depuis la première CPU et l'au moins une seconde CPU de la première topologie de CPU, et la déconnexion de la première CPU et de l'au moins une seconde CPU d'une CPU de la seconde topologie de CPU.
     
    7. Appareil d'ajout à chaud d'une unité centrale de traitement (CPU), dans lequel l'appareil est applicable à un serveur unique qui présente une première topologie de CPU à maillage incomplet, dans lequel l'appareil comprend :

    une unité de traitement (1101), configurée pour déterminer des premières informations d'indication, dans lequel les premières informations d'indication sont utilisées pour demander l'ajout d'une première CPU, et la première CPU n'est pas dans la première topologie de CPU en cours d'exploitation ; dans lequel

    l'unité de traitement (1101) est configurée en outre pour déterminer si au moins une seconde CPU est installée, dans lequel l'au moins une seconde CPU et la première CPU remplissent une condition prédéfinie ; et

    une unité d'envoi (1102), configurée pour : lorsque l'au moins une seconde CPU est installée, envoyer des secondes informations d'indication à la première topologie de CPU, dans lequel les secondes informations d'indication sont utilisées pour demander l'ajout de la première CPU et de l'au moins une seconde CPU, de manière à obtenir une seconde topologie de CPU et exploiter la seconde topologie de CPU ;

    dans lequel l'unité de traitement est configurée en outre pour :
    déterminer si au moins une seconde CPU de secours de la première CPU est installée, dans lequel l'au moins une seconde CPU de secours est l'au moins une seconde CPU.


     
    8. Appareil selon la revendication 7, comprenant en outre :

    une première unité de réception, configurée pour recevoir des premières informations d'indication en utilisant une interface utilisateur, dans lequel les premières informations d'indication comprennent un identifiant de la première CPU ; ou

    une seconde unité de réception, configurée pour recevoir, en utilisant un capteur, des secondes informations d'indication déclenchées lorsque la première CPU est installée ; dans lequel l'unité de réception est configurée en outre pour déterminer la première CPU installée selon les secondes informations d'indication.


     




    Drawing
































    Cited references

    REFERENCES CITED IN THE DESCRIPTION



    This list of references cited by the applicant is for the reader's convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard.

    Patent documents cited in the description