CN103516800A - Server system and operation system starting method and starting management node of server system - Google Patents

Server system and operation system starting method and starting management node of server system Download PDF

Info

Publication number
CN103516800A
CN103516800A CN201310456952.XA CN201310456952A CN103516800A CN 103516800 A CN103516800 A CN 103516800A CN 201310456952 A CN201310456952 A CN 201310456952A CN 103516800 A CN103516800 A CN 103516800A
Authority
CN
China
Prior art keywords
server
management node
pcie
port
startup management
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201310456952.XA
Other languages
Chinese (zh)
Other versions
CN103516800B (en
Inventor
牛功彪
于昌
刘钢
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
XFusion Digital Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to CN201310456952.XA priority Critical patent/CN103516800B/en
Publication of CN103516800A publication Critical patent/CN103516800A/en
Priority to PCT/CN2014/076217 priority patent/WO2015043175A1/en
Application granted granted Critical
Publication of CN103516800B publication Critical patent/CN103516800B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44Arrangements for executing specific programs
    • G06F9/4401Bootstrapping

Landscapes

  • Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Security & Cryptography (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Stored Programmes (AREA)

Abstract

The embodiment of the invention relates to the field of communication technology and discloses an operation system starting method and a server system based on the PCIE. The method comprises that a starting management node receives an access request sent by a first server through a PCIE network by utilizing a PCIE port; the starting management node guides the first server to have access to an operation system file of the first server to respond to the access request. According to the embodiment of the operation system starting method and the server system based on the PCIE, expenditure for starting a hard disc can be saved, meanwhile work load for replacing and maintaining the hard disc can be greatly reduced, and reliability of the server system is improved.

Description

Server system and os starting method thereof and startup management node
Technical field
The present invention relates to communication technical field, be specifically related to a kind of server system and os starting method thereof and start management node.
Background technology
In the scene of the application rack-mount servers such as data center, a server system comprises the server of One's name is legion, having a lot of situations is that every station server has oneself independently OS(Operating System, operating system) start hard disk, in the machine room of a lot of servers, hard disk is than the situation that is easier to occur inefficacy, so safeguarding hard disk and changing inefficacy hard disk has been exactly the important content of maintenance, because server sum is numerous, and the replacing of hard disk is frequent, and the workload of maintenance is very large.
Summary of the invention
Embodiment of the present invention technical problem to be solved is to provide a kind of server system and os starting method thereof and starts management node, for realizing single management node that starts, for a plurality of servers provide, starts service.
The embodiment of the present invention provides a kind of os starting method of server system, and described server system comprises high speed peripheral component interconnect Bus PC IE network, start management node and two-server at least; Wherein, described at least two-server is connected with described startup management node by described PCIE network, and described startup management node comprises at least one memory device, the operating system file of two-server at least described in described at least one memory device, stores has;
Described method comprises:
Described startup management node receives the access request that first server is used PCIE port to send by described PCIE network;
Described startup management node guides operating system file that described first server accesses described first server to respond described access request.
Accordingly, the embodiment of the present invention also provides a kind of os starting method of server system, and described server system comprises high speed peripheral component interconnect Bus PC IE network, start management node and two-server at least; Wherein, described at least two-server is connected with described startup management node by described PCIE network, described startup management node comprises at least one memory device, the operating system file of two-server at least described in described at least one memory device, stores has, and described method comprises:
First server is that high speed peripheral component interconnect Bus PC IE port starts by basic input-output system BIOS boot settings configuration;
Described first server is used PCIE port to described startup management node, to send access request by described PCIE network, and under the guiding of described startup management node, accesses the operating system file of described first server.
Accordingly, the embodiment of the present invention also provides a kind of startup management node, described startup management node is connected with two-server at least by high speed peripheral component interconnect Bus PC IE network, described startup management node comprises at least one memory device, the operating system file of two-server at least described in storing in described at least one memory device;
Described startup management node comprises:
Receiving element, the access request of using PCIE port to send by described PCIE network for receiving first server;
Guidance unit, for guiding operating system file that described first server accesses described first server to respond described access request.
Accordingly, the embodiment of the present invention also provides a kind of server system, and described server system comprises high speed peripheral component interconnect Bus PC IE network, foregoing startup management node and two-server at least.
The server system providing in the embodiment of the present invention and os starting method and startup management node, each server can start the startup that the startup management node of serving is realized operating system for multiple servers provides by PCIE access to netwoks, this locality at server just starts hard disk without configuration, can save the spending that starts hard disk, can reduce greatly the workload of hard disk replacement and maintenance, promote the reliability of server system simultaneously.
Accompanying drawing explanation
In order to be illustrated more clearly in the technical scheme in the embodiment of the present invention, to the accompanying drawing of required use in embodiment be briefly described below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skills, do not paying under the prerequisite of creative work, can also obtain according to these accompanying drawings other accompanying drawing.
Fig. 1 is flow process first schematic diagram of the os starting method of the server system that provides of embodiment mono-;
Fig. 2 is flow process second schematic diagram of the os starting method of the server system that provides of embodiment mono-;
Fig. 3 is flow process the 3rd schematic diagram of the os starting method of the server system that provides of embodiment mono-;
Fig. 4 is flow process first schematic diagram of the os starting method of the server system that provides of embodiment bis-;
Fig. 5 is flow process second schematic diagram of the os starting method of the server system that provides of embodiment bis-;
Fig. 6 is flow process the 3rd schematic diagram of the os starting method of the server system that provides of embodiment bis-
Fig. 7 is structure first schematic diagram of the startup management node that provides of embodiment tri-;
Fig. 8 is structure first schematic diagram of the startup management node that provides of embodiment tri-;
Fig. 9 is structure first schematic diagram of the startup management node that provides of embodiment tri-;
Figure 10 is the structural representation of the server system that provides of embodiment tetra-.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly described, obviously, described embodiment is only the present invention's part embodiment, rather than whole embodiment.Embodiment based in the present invention, those of ordinary skills, not making the every other embodiment obtaining under creative work prerequisite, belong to the scope of protection of the invention.
A kind of server system and os starting method are provided in the embodiment of the present invention and have started management node, for saving the spending of the startup hard disk of server system, can reduce greatly the workload of hard disk replacement and maintenance, promote the reliability of server system simultaneously.Below be elaborated respectively.
Term " first " in specification of the present invention and claims and above-mentioned accompanying drawing, " second ", " the 3rd " " 4th " etc. (if existence) are for distinguishing similar object, and needn't be for describing specific order or precedence.The data that should be appreciated that such use suitably can exchanged in situation, so as embodiments of the invention described herein for example can with except diagram here or describe those order enforcement.In addition, term " comprises " and " having " and their any distortion, intention is to cover not exclusive comprising, for example, those steps or unit that the process that has comprised series of steps or unit, method, system, product or equipment are not necessarily limited to clearly list, but can comprise clearly do not list or for these processes, method, product or equipment intrinsic other step or unit.
Embodiment mono-:
Fig. 1 is the schematic flow sheet of the os starting method of a kind of server system provided by the invention.As shown in Figure 1, the os starting method of this server system can comprise:
101, starting management node receives first server and uses PCIE(Peripheral Component Interface Express, high speed peripheral component interconnect bus) port is by the access request of above-mentioned PCIE network transmission;
Above-mentioned server system can comprise PCIE network and start management node and two-server at least; Wherein, above-mentioned at least two-server is connected with above-mentioned startup management node by PCIE network;
In the present embodiment, above-mentioned first server is to connect any one at least two-server that starts management node by PCIE network, and the starting method that the whole servers in above-mentioned at least two-server all can provide with the present embodiment starts the operating system;
In the present embodiment, start management node and can comprise at least one memory device, in this at least one memory device, can store the operating system file (as operating system mirror image) of above-mentioned at least two-server;
The quantity of this at least one memory device can be less than the quantity of above-mentioned at least two servers, has the situation of the shared memory device of a plurality of servers, can save compared to existing technology the quantity of memory device (startup hard disk);
102, above-mentioned startup management node guides operating system file that above-mentioned first server accesses above-mentioned first server to respond above-mentioned access request.
The os starting method utilization of the server system that the present embodiment provides starts management node provides os starting service for a plurality of servers, a plurality of servers just start hard disk without configuration in this locality, compared to existing technology, can save quantity and the spending of the startup hard disk of a plurality of servers needs, be convenient to start maintenance and the upgrading of hard disk, promote the reliability of server system.
In the present embodiment, above-mentioned memory device can be but be not limited to the memory devices such as hard disk, flash card, CD, USB flash disk, preferably above-mentioned memory device can be big capacity hard disk, as the startup hard disk that is above-mentioned at least two-server storage operation system file, take below and start hard disk and describe as example, but also above-mentioned memory device is not limited to start hard disk, and those skilled in the art also can select according to actual conditions other memory device.
Preferably, start management node and can only comprise a startup hard disk, can save to greatest extent the quantity that starts hard disk;
Preferably, in order to improve the reliability of single startup hard disk scheme, avoid because unique startup hard disk failure causes starting each server, start management node and also can comprise one or more standby startup hard disks.
Further, as shown in Figure 2, in some embodiments of the present invention, before above-mentioned 101, the method also can comprise:
201, above-mentioned startup management node is first port assignment the first memory space in above-mentioned at least one memory device; Wherein, above-mentioned the first memory space can be used for storing the operating system file of first server; Above-mentioned the first port is that above-mentioned first server connects the PCIE port of above-mentioned startup management node by above-mentioned PCIE network;
Take that to only have a memory device be example, the memory space of this memory device can be divided into N part (as being N district by hard disk partition), every part of memory space is used to the operating system file of this server of server stores, a memory device can be the startup service that N server provides operating system, and the numerical value of N can reasonably arrange according to the size of the amount of capacity of memory device and operating system file;
202, above-mentioned startup management node is mapped to above-mentioned the first memory space the PCIE address space of above-mentioned the first port;
203, above-mentioned startup management node loads management driving;
204, above-mentioned startup management node is that above-mentioned first server creates MBR(Master Boot Record, MBR at above-mentioned the first memory space), and above-mentioned MBR is labeled as to bootable startup.
Further, as shown in Figure 2, above-mentioned 102 can specifically comprise:
102A1, above-mentioned startup management node are resolved the address of the PCIE address space of above-mentioned the first port that above-mentioned access request comprises;
102B1, above-mentioned startup management node utilize the mapping relations of PCIE address space and above-mentioned first memory space of above-mentioned the first port, address by the address offset of the PCIE address space of above-mentioned the first port to above-mentioned the first memory space, so that above-mentioned first server is accessed the operating system file that above-mentioned the first memory space reads above-mentioned MBR and read above-mentioned first server under the guiding of above-mentioned MBR.
In execution mode as shown in Figure 2, start management node and complete the mapping to PCIE space to the first memory space, by port address, shine upon, first server can be accessed the first memory space according to mapping address, here the first memory space is just equivalent to the local hard drive of first server, thereby first server can read MBR and read operation system file starts the operating system.
In execution mode as shown in Figure 2, start management node and can complete the mapping to PCIE space to memory space corresponding to each server, by port address, shine upon, each server can be carried out leader record and start the operating system according to mapping address, eliminates the restriction that each server needs the startup storage of correspondence oneself.
Further, as shown in Figure 3, in some embodiments of the present invention, before above-mentioned 101, the method also can comprise:
301, above-mentioned startup management node is first port assignment the first memory space in above-mentioned at least one memory device; Wherein, above-mentioned the first memory space can be used for storing the operating system file of first server; Above-mentioned the first port is that above-mentioned first server connects the PCIE port of above-mentioned startup management node by above-mentioned PCIE network;
302, above-mentioned startup management node is the first port arrangement the one VF(Virtual Funtion, virtual functions);
303, above-mentioned startup management node configures an above-mentioned VF and a VHBA(Virtual Host Bus Adapters, virtual host bus adapter) mapping relations;
304, above-mentioned startup management node loads PF(Physical Function, physical function) driving;
305, above-mentioned startup management node loads management driving.
Further, as shown in Figure 3, above-mentioned 102 can specifically comprise:
102A2, above-mentioned startup management node are resolved the address of the above-mentioned VHBA that above-mentioned access request comprises;
102B2, above-mentioned startup management node are according to the mapping relations of an above-mentioned VHBA and an above-mentioned VF, by the address offset of an above-mentioned VHBA to above-mentioned YiVF address, so that above-mentioned first server is accessed an above-mentioned VF and utilize operating system file that the driving of an above-mentioned VF moves the above-mentioned first server of storing in above-mentioned the first memory space to above-mentioned first server.
Particularly, first server can be that operating system file is moved among the internal memory of above-mentioned first server, to improve the toggle speed of operating system.
In execution mode as shown in Figure 3, by VF being distributed to corresponding server, by starting the task of management node mapping memory space to PCIE space in Fig. 2, by hardware VF, completed completely, can improve the efficiency of execution.
It should be noted that, in execution mode as shown in Figure 3, starting management node needs to support SR-IOV(Single Root I/O Virtualization, single I/O Intel Virtualization Technology) technology.
Embodiment bis-:
Please refer to Fig. 4, Fig. 4 is the schematic flow sheet of the os starting method of the server system that provides of the embodiment of the present invention.As shown in Figure 4, the os starting method of this server system comprises:
401, first server is by BIOS(Basic Input Output System, basic input output system) boot settings configuration is that PCIE port starts;
Above-mentioned server system can comprise PCIE network and start management node and two-server at least; Wherein, above-mentioned at least two-server is connected with above-mentioned startup management node by PCIE network;
In the present embodiment, above-mentioned first server is to connect any one at least two-server that starts management node by PCIE network, and the starting method that the whole servers in above-mentioned at least two-server all can provide with the present embodiment starts the operating system;
402, above-mentioned first server is used PCIE port to above-mentioned startup management node, to send access request by above-mentioned PCIE network, and under the guiding of above-mentioned startup management node, accesses the operating system file of above-mentioned first server.
In the present embodiment, start management node and can comprise at least one memory device, in this at least one memory device, can store the operating system file (as operating system mirror image) of above-mentioned at least two-server;
Preferably, the quantity of this at least one memory device can be less than the quantity of above-mentioned at least two servers, has the situation of the shared memory device of a plurality of servers, can save compared to existing technology the quantity of memory device (startup hard disk).
As shown in Figure 5, in some embodiments of the present invention, above-mentioned 402 can comprise:
501, above-mentioned first server is used the first port to above-mentioned startup management node, to send access request by above-mentioned PCIE network, and above-mentioned access request comprises the address of the PCIE address space of above-mentioned the first port;
Above-mentioned the first port is that above-mentioned first server connects the PCIE port of above-mentioned startup management node by above-mentioned PCIE network;
502, above-mentioned first server receives the address of the first memory space that starts management node transmission, access above-mentioned the first memory space and read MBR MBR, and under the guiding of above-mentioned MBR, read the operating system file of the above-mentioned first server of storing in above-mentioned the first memory space;
The address of above-mentioned the first memory space utilizes the PCIE address space of above-mentioned the first port and the mapping relations of above-mentioned the first memory space to carry out address offset by above-mentioned startup management node and obtains.
As shown in Figure 6, in other execution modes of the present invention, above-mentioned 402 can comprise:
601, above-mentioned first server is used the first port to above-mentioned startup management node, to send access request by above-mentioned PCIE network, and above-mentioned access request comprises an above-mentioned VHBA(virtual host bus adapter) address; Above-mentioned the first port is that above-mentioned first server connects the PCIE port of above-mentioned startup management node by above-mentioned PCIE network;
602, above-mentioned first server receives the first virtual functions VF address that above-mentioned startup management node sends, access an above-mentioned VF, and utilize operating system file that the driving of an above-mentioned VF moves the above-mentioned first server of storing in the first memory space to above-mentioned first server;
Above-mentioned YiVF address can utilize the mapping relations of an above-mentioned VHBA and an above-mentioned VF to carry out address offset by above-mentioned startup management node and obtain.
In the server system os starting method providing in the present embodiment, each server can start the startup that the startup management node of serving is realized operating system for multiple servers provides by PCIE access to netwoks, this locality at server just starts hard disk without configuration, can save the spending that starts hard disk, can reduce greatly the workload of hard disk replacement and maintenance, promote the reliability of server system simultaneously.
Embodiment tri-:
Please refer to Fig. 7, Fig. 7 is a kind of structural representation that starts management node that the embodiment of the present invention provides.As shown in Figure 7, this startup management node 100 can comprise:
Receiving element 110, the access request of using PCIE port to send by above-mentioned PCIE network for receiving first server;
In the present embodiment, start management node 100 and be connected with two-server at least by PCIE network, form server system, above-mentioned first server is to connect any one at least two-server of startup management node by PCIE network;
Guidance unit 120, for guiding operating system file that above-mentioned first server accesses above-mentioned first server to respond above-mentioned access request.
Wherein, start management node 100 and comprise at least one memory device, in this at least one memory device, store the operating system file of above-mentioned at least two-server.
Preferably, the quantity of this at least one memory device can be less than the quantity of above-mentioned at least two servers, has the situation of the shared memory device of a plurality of servers, can save compared to existing technology the quantity of memory device (startup hard disk).
The startup management node that this enforcement provides can provide os starting service for two-server at least, each server can start the startup that management node is realized operating system by PCIE access to netwoks, this locality at server just starts hard disk without configuration, can save the spending that starts hard disk, can reduce greatly the workload of hard disk replacement and maintenance, promote the reliability of server system simultaneously.
In the present embodiment, above-mentioned memory device can be but be not limited to the memory devices such as hard disk, flash card, CD, USB flash disk, preferably above-mentioned memory device can be big capacity hard disk, startup hard disk as each server, take below and start hard disk and describe as example, but also above-mentioned memory device is not limited to start hard disk, and those skilled in the art also can select according to actual conditions other memory device.
Preferably, start management node 100 and can only comprise a startup hard disk, can save to greatest extent the quantity that starts hard disk;
Preferably, in order to improve the reliability of single startup hard disk scheme, avoid because unique startup hard disk failure causes starting each server, start management node 100 and also can comprise one or more standby startup hard disks.
Further, in some embodiments of the present invention, as shown in Figure 8, start management node 100 and also can comprise:
The first allocation units 130, for being first port assignment the first memory space at above-mentioned at least one memory device, wherein above-mentioned the first memory space can be used for storing the operating system file of first server; Above-mentioned the first port is that above-mentioned first server connects the PCIE port that starts management node by above-mentioned PCIE network;
Map unit 140, for being mapped to above-mentioned the first memory space the PCIE address space of above-mentioned the first port;
The first drive load unit 150, the management that starts management node 100 for loading drives;
Creating unit 160, for being that first server creates MBR at above-mentioned the first memory space, and is labeled as bootable startup by above-mentioned MBR.
Further, in execution mode as shown in Figure 8, guidance unit 120 can comprise:
First resolves subelement, for resolving the address of the PCIE address space of above-mentioned the first port that above-mentioned access request comprises;
The first address offset subelement, be used for according to the mapping relations of the PCIE address space of above-mentioned the first port and above-mentioned the first memory space, address by the address offset of the PCIE address space of above-mentioned the first port to above-mentioned the first memory space, so that above-mentioned first server is accessed the operating system file that above-mentioned the first memory space reads above-mentioned MBR and read above-mentioned first server under the guiding of above-mentioned MBR.
In execution mode as shown in Figure 8, start management node 100 and complete the mapping to PCIE space to the first memory space, by port address, shine upon, first server can be accessed the first memory space according to mapping address, here the first memory space is just equivalent to the local hard drive of first server, and first server can read MBR and start the operating system.
In execution mode as shown in Figure 8, start management node 100 and can complete the mapping to PCIE space to memory space corresponding to each server, by port address, shine upon, each server can be carried out leader record and start the operating system according to mapping address, eliminates the restriction that each server needs the startup storage of correspondence oneself.
Further, in some embodiments of the present invention, as shown in Figure 9, start management node 100 and also can comprise:
The second allocation units 170, for being first port assignment the first memory space at above-mentioned at least one memory device, above-mentioned the first memory space is for storing the operating system file of above-mentioned first server; Above-mentioned the first port is the PCIE port that above-mentioned first server connects above-mentioned PCIE network;
The first dispensing unit 180, is used to the first port arrangement the one VF, and configures the mapping relations of an above-mentioned VF and above-mentioned the first memory space; Above-mentioned the first port is the PCIE port that above-mentioned first server connects above-mentioned PCIE network;
The second dispensing unit 190, for configuring the mapping relations of an above-mentioned VF and a VHBA;
The second drive load unit 1A0, for loading the driving of physical function PF;
The 3rd drive load unit 1B0, drives for loading the management of above-mentioned startup management node.
Further, in execution mode as shown in Figure 9, guidance unit 120 can specifically comprise:
Second resolves subelement, for resolving the address of the above-mentioned VHBA that above-mentioned access request comprises;
The second address offset subelement, be used for according to the mapping relations of an above-mentioned VHBA and an above-mentioned VF, by the address offset of an above-mentioned VHBA to above-mentioned YiVF address, so that above-mentioned first server access the one VF utilize operating system file that the driving of an above-mentioned VF moves the above-mentioned first server of storing in above-mentioned the first memory space to above-mentioned first server.
It should be noted that, in execution mode as shown in Figure 9, start management node 100 and need to support SR-IOV technology.In execution mode as shown in Figure 9, by VF being distributed to corresponding server, by starting the task of management node mapping memory space to PCIE space in Fig. 8, by hardware VF, completed completely, can improve the efficiency of execution.
Embodiment tetra-:
Please refer to Figure 10, Figure 10 is the structural representation of a kind of server system of providing of the embodiment of the present invention.As shown in figure 10, this server system comprises PCIE network and starts management node and two-server at least, wherein, this at least two-server by PCIE network, be connected with above-mentioned startup management node, this startup management node can be startup management node 100 as described in embodiment tri-, particular content can reference implementation three description, do not repeat them here.
Each embodiment is all based on same inventive concept above, and the description of each embodiment emphasizes particularly on different fields, and in indivedual embodiment, describes not detailed part, can be with reference to the description in other embodiment.
One of ordinary skill in the art will appreciate that all or part of step in the whole bag of tricks of above-described embodiment is to come the hardware that instruction is relevant to complete by program, this program can be stored in a computer-readable recording medium, storage medium can comprise: flash disk, read-only memory (Read-Only Memory, ROM), random access device (Random Access Memory, RAM), disk or CD etc.
The server system above embodiment of the present invention being provided and os starting method thereof and startup management node are described in detail, applied specific case herein principle of the present invention and execution mode are set forth, the explanation of above embodiment is just for helping to understand method of the present invention and core concept thereof; , for one of ordinary skill in the art, according to thought of the present invention, all will change in specific embodiments and applications, in sum, this description should not be construed as limitation of the present invention meanwhile.

Claims (14)

1. an os starting method for server system, is characterized in that, described server system comprises high speed peripheral component interconnect Bus PC IE network, start management node and two-server at least; Wherein, described at least two-server is connected with described startup management node by described PCIE network, and described startup management node comprises at least one memory device, the operating system file of two-server at least described in described at least one memory device, stores has;
Described method comprises:
Described startup management node receives the access request that first server is used PCIE port to send by described PCIE network;
Described startup management node guides operating system file that described first server accesses described first server to respond described access request.
2. method according to claim 1, is characterized in that, before described startup management node receives the access request of first server use PCIE port by described PCIE network transmission, described method also comprises:
Described startup management node is first port assignment the first memory space in described at least one memory device, and described the first memory space is for storing the operating system file of described first server; Described the first port is that described first server connects the PCIE port of described startup management node by described PCIE network;
Described startup management node is mapped to described the first memory space the PCIE address space of described the first port;
Described startup management node loads management and drives;
Described startup management node is that described first server creates MBR MBR at described the first memory space, and described MBR is labeled as to bootable startup.
3. method according to claim 2, is characterized in that, the operating system file that described startup management node guides described first server to access described first server comprises to respond described access request:
Described startup management node is resolved the address of the PCIE address space of described the first port that described access request comprises;
Described startup management node utilizes the mapping relations of PCIE address space and described first memory space of described the first port, address by the address offset of the PCIE address space of described the first port to described the first memory space, so that described first server is accessed the operating system file that described the first memory space reads described MBR and read described first server under the guiding of described MBR.
4. method according to claim 1, is characterized in that, before described startup management node receives the access request of first server use PCIE port by described PCIE network transmission, described method also comprises:
Described startup management node is first port assignment the first memory space in described at least one memory device, and described the first memory space is for storing the operating system file of described first server; Described the first port is the PCIE port that described first server connects described PCIE network;
Described startup management node is described first port arrangement the first virtual functions VF, and configures the mapping relations of a described VF and described the first memory space;
Described startup management node configures the mapping relations of a described VF and the first virtual host bus adapter VHBA;
Described startup management node loads the driving of physical function PF;
Described startup management node loads management and drives.
5. method according to claim 4, is characterized in that, the operating system file that described startup management node guides described first server to access described first server comprises to respond described access request:
Described startup management node is resolved the address of the described VHBA that described access request comprises;
Described startup management node is according to the mapping relations of a described VHBA and a described VF, by the address offset of a described VHBA to described YiVF address, so that described first server is accessed a described VF and utilize operating system file that the driving of a described VF moves the described first server of storing in described the first memory space to described first server.
6. an os starting method for server system, is characterized in that, described server system comprises high speed peripheral component interconnect Bus PC IE network, start management node and two-server at least; Wherein, described at least two-server is connected with described startup management node by described PCIE network, described startup management node comprises at least one memory device, the operating system file of two-server at least described in described at least one memory device, stores has, and described method comprises:
First server is that PCIE port starts by basic input-output system BIOS boot settings configuration;
Described first server is used PCIE port to described startup management node, to send access request by described PCIE network, and under the guiding of described startup management node, accesses the operating system file of described first server.
7. method according to claim 6, it is characterized in that, described first server use PCIE port by described PCIE network to described startup management node transmission access request, and the operating system file of accessing described first server under the guiding of described startup management node comprises:
Described first server is used the first port to described startup management node, to send access request by described PCIE network, and described access request comprises the address of the PCIE address space of described the first port; Described the first port is that described first server connects the PCIE port of described startup management node by described PCIE network;
Described first server receives the address of the first memory space that starts management node transmission, access described the first memory space and read MBR MBR, and under the guiding of described MBR, read the operating system file of the described first server of storing in described the first memory space; The address of described the first memory space utilizes the PCIE address space of described the first port and the mapping relations of described the first memory space to carry out address offset by described startup management node and obtains.
8. method according to claim 6, it is characterized in that, described first server use PCIE port by described PCIE network to described startup management node transmission access request, and the operating system file of accessing described first server under the guiding of described startup management node comprises:
Described first server is used the first port to described startup management node, to send access request by described PCIE network, and described access request comprises the address of described the first virtual host bus adapter VHBA; Described the first port is that described first server connects the PCIE port of described startup management node by described PCIE network;
Described first server receives the first virtual functions VF address that described startup management node sends, access a described VF, and utilize operating system file that the driving of a described VF moves the described first server of storing in the first memory space to described first server; Described YiVF address utilizes the mapping relations of a described VHBA and a described VF to carry out address offset by described startup management node and obtains.
9. one kind starts management node, it is characterized in that, described startup management node is connected with two-server at least by high speed peripheral component interconnect Bus PC IE network, described startup management node comprises at least one memory device, the operating system file of two-server at least described in storing in described at least one memory device;
Described startup management node comprises:
Receiving element, the access request of using PCIE port to send by described PCIE network for receiving first server;
Guidance unit, for guiding operating system file that described first server accesses described first server to respond described access request.
10. startup management node according to claim 9, is characterized in that, described startup management node also comprises:
The first allocation units, for being first port assignment the first memory space at described at least one memory device, described the first memory space is for storing the operating system file of described first server; Described the first port is that described first server connects the PCIE port of described startup management node by described PCIE network;
Map unit, for being mapped to described the first memory space the PCIE address space of described the first port;
The first drive load unit, drives for loading the management of described startup management node;
Creating unit, for being that described first server creates MBR MBR at described the first memory space, and is labeled as bootable startup by described MBR.
11. startup management nodes according to claim 10, is characterized in that, described guidance unit comprises:
First resolves subelement, for resolving the address of the PCIE address space of described the first port that described access request comprises;
The first address offset subelement, be used for according to the mapping relations of the PCIE address space of described the first port and described the first memory space, address by the address offset of the PCIE address space of described the first port to described the first memory space, so that described first server is accessed the operating system file that described the first memory space reads described MBR and read described first server under the guiding of described MBR.
12. startup management nodes according to claim 9, is characterized in that, described startup management node also comprises:
The second allocation units, for being first port assignment the first memory space at described at least one memory device, described the first memory space is for storing the operating system file of described first server; Described the first port is the PCIE port that described first server connects described PCIE network;
The first dispensing unit, is used to first port arrangement the first virtual functions VF, and configures the mapping relations of a described VF and described the first memory space;
The second dispensing unit, for configuring the mapping relations of a described VF and the first virtual host bus adapter VHBA;
The 3rd drive load unit, for loading the driving of physical function PF;
4 wheel driven moves loading unit, for loading management, drives.
13. startup management nodes according to claim 12, is characterized in that, described guidance unit comprises:
Second resolves subelement, for resolving the address of the described VHBA that described access request comprises;
The second address offset subelement, be used for according to the mapping relations of a described VHBA and a described VF, by the address offset of a described VHBA to described YiVF address, so that described first server is accessed a described VF and utilize operating system file that the driving of a described VF moves the described first server of storing in described the first memory space to described first server.
14. 1 kinds of server systems, is characterized in that, described server system comprises at least two-server, high speed peripheral component interconnect Bus PC IE network and the startup management node as described in claim 9-13 any one.
CN201310456952.XA 2013-09-29 2013-09-29 Server system and its os starting method and startup management node Active CN103516800B (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201310456952.XA CN103516800B (en) 2013-09-29 2013-09-29 Server system and its os starting method and startup management node
PCT/CN2014/076217 WO2015043175A1 (en) 2013-09-29 2014-04-25 Server system and operation system starting method thereof, and starting management node

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310456952.XA CN103516800B (en) 2013-09-29 2013-09-29 Server system and its os starting method and startup management node

Publications (2)

Publication Number Publication Date
CN103516800A true CN103516800A (en) 2014-01-15
CN103516800B CN103516800B (en) 2018-04-10

Family

ID=49898817

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310456952.XA Active CN103516800B (en) 2013-09-29 2013-09-29 Server system and its os starting method and startup management node

Country Status (2)

Country Link
CN (1) CN103516800B (en)
WO (1) WO2015043175A1 (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2015043175A1 (en) * 2013-09-29 2015-04-02 华为技术有限公司 Server system and operation system starting method thereof, and starting management node
CN105068836A (en) * 2015-08-06 2015-11-18 北京百度网讯科技有限公司 SAS (serial attached SCSI) network based remotely-shareable start-up system
CN108604115A (en) * 2015-12-21 2018-09-28 国际商业机器公司 Distributed operating system function for the node in rack

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101206633A (en) * 2006-12-19 2008-06-25 国际商业机器公司 System and method for communication between host systems using a transaction protocol and shared memories
CN101770345A (en) * 2008-12-31 2010-07-07 成都市华为赛门铁克科技有限公司 Method for establishing logical address space, method for access to storage device and storage architecture
CN101989212A (en) * 2009-07-31 2011-03-23 国际商业机器公司 Method and device for providing virtual machine management program for starting blade server
US20120079143A1 (en) * 2010-09-24 2012-03-29 Xsigo Systems, Inc. Wireless host i/o using virtualized i/o controllers
CN102609215A (en) * 2012-04-11 2012-07-25 成都市华为赛门铁克科技有限公司 Data processing method and device

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101299757A (en) * 2008-05-23 2008-11-05 华为技术有限公司 Data sharing method and communication system as well as correlation equipment
CN103516800B (en) * 2013-09-29 2018-04-10 华为技术有限公司 Server system and its os starting method and startup management node

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101206633A (en) * 2006-12-19 2008-06-25 国际商业机器公司 System and method for communication between host systems using a transaction protocol and shared memories
CN101770345A (en) * 2008-12-31 2010-07-07 成都市华为赛门铁克科技有限公司 Method for establishing logical address space, method for access to storage device and storage architecture
CN101989212A (en) * 2009-07-31 2011-03-23 国际商业机器公司 Method and device for providing virtual machine management program for starting blade server
US20120079143A1 (en) * 2010-09-24 2012-03-29 Xsigo Systems, Inc. Wireless host i/o using virtualized i/o controllers
CN102609215A (en) * 2012-04-11 2012-07-25 成都市华为赛门铁克科技有限公司 Data processing method and device

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2015043175A1 (en) * 2013-09-29 2015-04-02 华为技术有限公司 Server system and operation system starting method thereof, and starting management node
CN105068836A (en) * 2015-08-06 2015-11-18 北京百度网讯科技有限公司 SAS (serial attached SCSI) network based remotely-shareable start-up system
CN108604115A (en) * 2015-12-21 2018-09-28 国际商业机器公司 Distributed operating system function for the node in rack
CN108604115B (en) * 2015-12-21 2021-05-28 国际商业机器公司 Distributed operating system functionality for nodes in a chassis

Also Published As

Publication number Publication date
CN103516800B (en) 2018-04-10
WO2015043175A1 (en) 2015-04-02

Similar Documents

Publication Publication Date Title
CN110704155B (en) Container network construction method and device, physical host and data transmission method
US9531668B2 (en) Micro server, method of allocating MAC address, and computer readable recording medium
WO2017152633A1 (en) Port binding implementation method and device
CN103210379A (en) Server system, management method and device
US10372639B2 (en) System and method to avoid SMBus address conflicts via a baseboard management controller
US20220353237A1 (en) Systems and methods of database instance container deployment
CN104657305A (en) Method and system for implementing dynamic virtualization of an SRIOV capable SAS adapter
CN104461958A (en) Storage resource access method supporting SR-IOV, storage controller and storage equipment
CN103890728A (en) Method for hot transfer of virtual machine and server
US20170031698A1 (en) System and method for dynamically allocating resources of virtual machines based on service-level agreements (sla) and privilege levels of users
US9336032B2 (en) Zoning data to a virtual machine
CN103501242A (en) Resource management method and multiple-node cluster device
CN107395710B (en) Method and device for realizing configuration and high availability HA of cloud platform network element
CN105335168A (en) System, method and apparatus for remotely configuring operating system
CN102316043A (en) Port virtualization method, switch and communication system
US9280493B2 (en) Method and device for enumerating input/output devices
US10719333B2 (en) BIOS startup method and apparatus
JP2023502673A (en) Virtual drawer in server
CN103516800A (en) Server system and operation system starting method and starting management node of server system
CN103530254A (en) Peripheral Component Interconnect enumeration method and device of multi-node system
TWI616759B (en) Apparatus assigning controller and apparatus assigning method
CN104123188A (en) Resource allocating method and related devices
US10877918B2 (en) System and method for I/O aware processor configuration
CN102393838A (en) Data processing method and device, PCI-E (peripheral component interface-express) bus system, and server
CN104104594A (en) Methods of transmitting and receiving protocol message of VSU, devices and system

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right

Effective date of registration: 20211221

Address after: 450046 Floor 9, building 1, Zhengshang Boya Plaza, Longzihu wisdom Island, Zhengdong New Area, Zhengzhou City, Henan Province

Patentee after: Super fusion Digital Technology Co.,Ltd.

Address before: 518129 Bantian HUAWEI headquarters office building, Longgang District, Guangdong, Shenzhen

Patentee before: HUAWEI TECHNOLOGIES Co.,Ltd.

TR01 Transfer of patent right