CN104965571A - Server - Google Patents
Server Download PDFInfo
- Publication number
- CN104965571A CN104965571A CN201510320856.1A CN201510320856A CN104965571A CN 104965571 A CN104965571 A CN 104965571A CN 201510320856 A CN201510320856 A CN 201510320856A CN 104965571 A CN104965571 A CN 104965571A
- Authority
- CN
- China
- Prior art keywords
- pcie
- data
- cpu
- server
- subcard
- 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
Links
Abstract
The present invention provides a server which comprises at least two PCIE slots, a PCIE switching apparatus, at least two connectors, a network switching apparatus and a management CPU, wherein a PCIE switching chip is also respectively connected with each PCIE slot and an external upper-layer server so as to form a first data link; and the at least two PCIE slots are formed correspondingly to the at least two connectors, each connector is connected to the network switching apparatus and the network switching apparatus is also connected with the management CPU so as to form a second data link. According to the scheme, various types of data processing can be carried out.
Description
Technical field
The present invention relates to network communication technology field, particularly a kind of server.
Background technology
In the computing machine in modern times, the fan of particularly domestic system and game, the process of figure becomes more and more important, need the core processor of a special figure, therefore, graphic process unit (GPU, Graphic Processing Unit) is arisen at the historic moment.Along with the development of GPU, GPU is no longer confined to graphics process, the concern that the technical development of GPU general-purpose computations has caused industry many, the fact also proves in the calculating of the part such as floating-point operation, parallel computation, and even GPU can provide decades of times hundreds of times in the performance of CPU.
At present, in the server comprising GPU card, carry out data processing by GPU card, and send to the server on upper strata.Owing to only utilizing GPU card to carry out data processing, therefore, processing mode is single, cannot meet user's multiple business demand.
Summary of the invention
The invention provides a kind of server, several data processing mode can be realized.
A kind of server, comprising:
At least two PCIE slots, PCIE switch, at least two connectors, network-switching equipment, CPU management, wherein,
Described CPU management, is connected with described PCIE switch;
Described PCIE switch is also connected with each PCIE slot and outside top service device respectively, to form the first data link;
Described at least two PCIE slots and described at least two connector one_to_one corresponding are arranged, and each connector is all connected to described network-switching equipment, and described network-switching equipment is also connected with described CPU management, to form the second data link.
Described CPU management is safeguarded described PCIE switch.
Comprise further:
Be plugged at least one the GPU card at least one PCIE slot;
And/or,
Be plugged at least one the CPU subcard at least one PCIE slot.
Described CPU subcard comprises connecting interface, and each CPU subcard is connected by the connector that connecting interface is corresponding with the PCIE slot of institute grafting.
In described first data link,
Data are transferred to described PCIE switch by PCIE signal line by the GPU card on described PCIE slot, and the data that GPU card is sent are sent to outside top service device by described PCIE switch;
Described PCIE interactive device receives the data that outside top service device is sent, and is transferred to the GPU card on corresponding PCIE slot by PCIE signal line.
In described first data link,
Data are transferred to described PCIE switch by PCIE signal line by the CPU subcard on described PCIE slot, and the data that CPU subcard is sent are sent to outside top service device by described PCIE switch;
Described PCIE switch receives the data that outside top service device is sent, and is transferred to the CPU subcard on corresponding PCIE slot by PCIE signal line.
In described second data link,
CPU subcard on described PCIE slot is by connecting interface and connected connector, and send the data to described network-switching equipment, the data of CPU subcard are sent to described CPU management by described network-switching equipment;
Described CPU management sends the data to described network-switching equipment, and described network-switching equipment sends the data to the CPU subcard on corresponding PCIE slot by corresponding connector.
Comprise further: four boards, wherein, two boards are perpendicular slotting at server internal, and another two boards are in server internal horizontal positioned; Further,
Described CPU management, PCIE switch and network-switching equipment are all arranged on described perpendicular two boards inserted;
Described PCIE slot is arranged on two boards of horizontal positioned.
Comprise 16 described PCIE slots.
Described server is the GPU server of 4U.
Embodiments provide a kind of server, this server comprises the first data link and second data link of the formation of each hardware, wherein, the PCIE switch that first data link is corresponding to be connected with outside top service device and the PCIE slot of pluggable GPU card, thus carry out the transmission of GPU card data; Second data link the is corresponding PCIE slot of CPU management, network-switching equipment, connector and pluggable CPU subcard, thus carry out the transmission of data between CPU management and CPU subcard.Therefore, achieve several data processing mode, same server-compatible PCIE exchanges and network exchange.
Accompanying drawing explanation
Fig. 1 is a kind of structural representation of server in one embodiment of the invention.
Fig. 2 is a kind of structural representation of server in another embodiment of the present invention.
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 and completely described.Obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.
One embodiment of the invention proposes a kind of server, see Fig. 1, comprising:
At least two PCIE slot 101, PCIE switches 102, at least two connectors 103, network-switching equipment 104, CPU management 105, wherein,
Described CPU management 105, is connected with described PCIE switch 102;
Described PCIE switch 102 is also connected with each PCIE slot 101 and outside top service device respectively, to form the first data link;
Described at least two PCIE slots 101 are arranged with described at least two connectors 103 one_to_one corresponding, and each connector 103 is all connected to described network-switching equipment 104, described network-switching equipment 104 is also connected with described CPU management 105, to form the second data link.
Visible, the first data link and second data link of the formation of each hardware is comprised at the server of embodiment of the present invention proposition, wherein, the PCIE switch that first data link is corresponding to be connected with outside top service device and the PCIE slot of pluggable GPU card, thus carry out the transmission of GPU card data; Second data link the is corresponding PCIE slot of CPU management, network-switching equipment, connector and pluggable CPU subcard, thus carry out the transmission of data between CPU management and CPU subcard.Therefore, achieve several data processing mode, same server-compatible PCIE exchanges and network exchange.
In the server shown in above-mentioned Fig. 1, described CPU management 105 can be safeguarded described PCIE switch 104, such as carries out the quantity etc. of GPU card and CPU subcard on power on, running status is safeguarded, PCIE switch connects PCIE slot.
See Fig. 2, in an embodiment of the invention, multiple PCIE slot 101 is plugged with GPU card 201 and CPU subcard 202 simultaneously.Such as, always have 16 PCIE slots 101, wherein, 10 PCIE slots 101 are plugged with GPU card, another 6 PCIE slots 101 are plugged with CPU subcard.
Certainly, in the service application of reality, also can on all PCIE slots all grafting GPU cards, or, all grafting CPU subcard on all PCIE slots, thus meet the various demands of business.
In an embodiment of the invention, any one CPU subcard can comprise connecting interface, each CPU subcard is connected by the connector that connecting interface is corresponding with the PCIE slot of institute grafting.
Any one server utilizing the embodiment of the present invention to propose, can realize several data processing mode.For above-mentioned Fig. 2, the possible working method of prescribed server:
Working method one: when being plugged with GPU card on PCIE slot, in described first data link, can realize the transmission of GPU card data, comprising:
Data are transferred to described PCIE switch 102 by PCIE signal line by the GPU card 201 on described PCIE slot 101, and the data that GPU card 201 is sent are sent to outside top service device by described PCIE switch 102; Further,
Described PCIE switch 102 receives the data that outside top service device is sent, and is transferred to the GPU card 201 on corresponding PCIE slot 101 by PCIE signal line.
Working method two: when being plugged with CPU subcard on PCIE slot, in described first data link, can realize the transmission of CPU subcard data, comprising:
In described first data link,
Data are transferred to described PCIE switch 102 by PCIE signal line by the CPU subcard 202 on described PCIE slot 101, and the data that CPU subcard 202 is sent are sent to outside top service device by described PCIE switch 102;
Described PCIE switch 102 receives the data that outside top service device is sent, and is transferred to the CPU subcard 202 on corresponding PCIE slot 101 by PCIE signal line.
Working method three: when being plugged with CPU subcard on PCIE slot, in described second data link, can realize the transmission of data between CPU management and CPU subcard,
In described second data link,
CPU subcard 202 on described PCIE slot 101 is by connecting interface and connected connector 103, and send the data to described network-switching equipment 104, the data of CPU subcard 202 are sent to described CPU management 105 by described network-switching equipment 104;
Described CPU management 105 sends the data to described network-switching equipment 104, and described network-switching equipment 104 sends the data to the CPU subcard 202 on corresponding PCIE slot 101 by corresponding connector 103.
In an embodiment of the invention, may further include in server: four boards, wherein, two boards are perpendicular slotting at server internal, and another two boards are in server internal horizontal positioned; Further,
Described CPU management, PCIE switch and network-switching equipment are all arranged on described perpendicular two boards inserted; Described PCIE slot is arranged on two boards of horizontal positioned.
In an embodiment of the invention, server internal comprises 16 PCIE slots.
In an embodiment of the invention, described server is the GPU server of 4U, that is, can realize two data link to the GPU server of 4U, and compatible with PCI E exchanges and network exchange.
Each embodiment of the present invention at least has following beneficial effect:
1, server comprises the first data link and second data link of each hardware formation, wherein, the PCIE switch that first data link is corresponding to be connected with outside top service device and the PCIE slot of pluggable GPU card, thus carry out the transmission of GPU card data; Second data link the is corresponding PCIE slot of CPU management, network-switching equipment, connector and pluggable CPU subcard, thus carry out the transmission of data between CPU management and CPU subcard.Therefore, achieve several data processing mode, same server-compatible PCIE exchanges and network exchange.
2, due to can grafting CPU subcard on PCIE slot, the data realized between CPU subcard and CPU management by the second data link be transmitted, therefore, it is possible to realize multiple application, such as, CPU subcard is to the load balancing of CPU management.
3, at least two PCIE slots, can according to service needed, whole grafting GPU card, or, omnidistance grafting CPU subcard, or, part grafting GPU card and part grafting CPU subcard, therefore, it is possible to according to practical business need carry out flexible configuration.
4, in embodiments of the invention, also the data of CPU subcard can be passed through the first data link transmission to outside top service device, namely, the data that CPU subcard on PCIE slot and PCIE switch can carry out CPU subcard are transmitted, so that outside top service device communication, thus further expand the dirigibility of business realizing.
It should be noted that, in this article, the relational terms of such as first and second and so on is only used for an entity or operation to separate with another entity or operational zone, and not necessarily requires or imply the relation that there is any this reality between these entities or operation or sequentially.And, term " comprises ", " comprising " or its any other variant are intended to contain comprising of nonexcludability, thus make to comprise the process of a series of key element, method, article or equipment and not only comprise those key elements, but also comprise other key elements clearly do not listed, or also comprise by the intrinsic key element of this process, method, article or equipment.When not more restrictions, the key element limited by statement " comprising ... ", and be not precluded within process, method, article or the equipment comprising described key element and also there is other same factor.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, within the spirit and principles in the present invention all, any amendment made, equivalent replacement, improvement etc., all should be included within the scope of protection of the invention.
Claims (10)
1. a server, is characterized in that, comprising:
At least two PCIE slots, PCIE switch, at least two connectors, network-switching equipment, CPU management, wherein,
Described PCIE exchange chip is also connected with each PCIE slot and outside top service device respectively, to form the first data link;
Described at least two PCIE slots and described at least two connector one_to_one corresponding are arranged, and each connector is all connected to described network-switching equipment, and described network-switching equipment is also connected with described CPU management, to form the second data link.
2. server according to claim 1, is characterized in that, described CPU management, is also connected with described PCIE switch, safeguards described PCIE switch.
3. server according to claim 1, is characterized in that, comprises further:
Be plugged at least one the GPU card at least one PCIE slot;
And/or,
Be plugged at least one the CPU subcard at least one PCIE slot.
4. server according to claim 3, is characterized in that, described CPU subcard comprises connecting interface, and each CPU subcard is connected by the connector that connecting interface is corresponding with the PCIE slot of institute grafting.
5. server according to claim 3, is characterized in that, in described first data link,
Data are transferred to described PCIE interactive device by PCIE signal line by the GPU card on described PCIE slot, and the data that GPU card is sent are sent to outside top service device by described PCIE switch;
Described PCIE interactive device receives the data that outside top service device is sent, and is transferred to the GPU card on corresponding PCIE slot by PCIE signal line.
6. server according to claim 5, is characterized in that, in described first data link,
Data are transferred to described PCIE switch by PCIE signal line by the CPU subcard on described PCIE slot, and the data that CPU subcard is sent are sent to outside top service device by described PCIE interactive device;
Described PCIE switch receives the data that outside top service device is sent, and is transferred to the CPU subcard on corresponding PCIE slot by PCIE signal line.
7. server according to claim 3, is characterized in that, in described second data link,
CPU card on described PCIE slot is by described connecting interface and connected connector, and send the data to described network-switching equipment, the data of CPU subcard are sent to described CPU management by described network-switching equipment;
Described CPU management sends the data to described network-switching equipment, and described network-switching equipment sends the data to the CPU subcard on corresponding PCIE slot by corresponding connector.
8. according to described server arbitrary in claim 1 to 7, it is characterized in that, comprise further: four boards, wherein, two boards are perpendicular slotting at server internal, and another two boards are in server internal horizontal positioned; Further,
Described CPU management, PCIE switch and network-switching equipment are all arranged on described perpendicular two boards inserted;
Described PCIE slot is arranged on two boards of horizontal positioned.
9. server according to claim 8, is characterized in that, comprises 16 described PCIE slots.
10., according to described server arbitrary in claim 1 to 7, it is characterized in that, described server is the GPU server of 4U.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510320856.1A CN104965571B (en) | 2015-06-11 | 2015-06-11 | A kind of server |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510320856.1A CN104965571B (en) | 2015-06-11 | 2015-06-11 | A kind of server |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104965571A true CN104965571A (en) | 2015-10-07 |
| CN104965571B CN104965571B (en) | 2018-05-25 |
Family
ID=54219609
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510320856.1A Active CN104965571B (en) | 2015-06-11 | 2015-06-11 | A kind of server |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104965571B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107748726A (en) * | 2017-11-02 | 2018-03-02 | 郑州云海信息技术有限公司 | A kind of GPU casees |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080307149A1 (en) * | 2007-06-08 | 2008-12-11 | Tomonori Hirai | Clustering System and Flexible Interconnection Architecture Thereof |
| CN103036817A (en) * | 2012-12-14 | 2013-04-10 | 华为技术有限公司 | Server single-board, server single-board realization method and host processor |
| CN203561931U (en) * | 2013-11-22 | 2014-04-23 | 曙光信息产业(北京)有限公司 | External GPU card extending device |
| CN103793003A (en) * | 2012-10-31 | 2014-05-14 | 华为技术有限公司 | Exchange plate and blade server |
| CN204044694U (en) * | 2014-08-27 | 2014-12-24 | 浪潮电子信息产业股份有限公司 | A kind of low cost expanded type GPU blade server |
-
2015
- 2015-06-11 CN CN201510320856.1A patent/CN104965571B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080307149A1 (en) * | 2007-06-08 | 2008-12-11 | Tomonori Hirai | Clustering System and Flexible Interconnection Architecture Thereof |
| CN103793003A (en) * | 2012-10-31 | 2014-05-14 | 华为技术有限公司 | Exchange plate and blade server |
| CN103036817A (en) * | 2012-12-14 | 2013-04-10 | 华为技术有限公司 | Server single-board, server single-board realization method and host processor |
| CN203561931U (en) * | 2013-11-22 | 2014-04-23 | 曙光信息产业(北京)有限公司 | External GPU card extending device |
| CN204044694U (en) * | 2014-08-27 | 2014-12-24 | 浪潮电子信息产业股份有限公司 | A kind of low cost expanded type GPU blade server |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107748726A (en) * | 2017-11-02 | 2018-03-02 | 郑州云海信息技术有限公司 | A kind of GPU casees |
| CN107748726B (en) * | 2017-11-02 | 2020-03-24 | 郑州云海信息技术有限公司 | GPU (graphics processing Unit) box |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104965571B (en) | 2018-05-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102841638B (en) | Design method of multifunctional VPX back panel | |
| KR200466902Y1 (en) | High integration computer system | |
| CN102866729A (en) | server cabinet system | |
| CN104915917A (en) | GPU cabinet, PCIe exchange device and server system | |
| CN107480094A (en) | A kind of pond server system architecture of fusion architecture | |
| CN104641593B (en) | Web plate and communication equipment | |
| CN104932618A (en) | GPU (Graphics Processing Unit) server equipment | |
| CN109410117B (en) | Graphics processor system | |
| CN104965571A (en) | Server | |
| CN203133718U (en) | Blade server backboard | |
| CN102223239B (en) | Exchange card and rack type equipment | |
| CN205091663U (en) | Modularization server backplate based on CPCI framework | |
| CN105095143A (en) | Server node and complete machine cabinet server | |
| CN106774700A (en) | A kind of rack-mount server | |
| CN106161169A (en) | A kind of multi-host network exchange system | |
| CN206312032U (en) | A kind of multi-functional main frame | |
| CN104994040A (en) | Ethernet switch and port multiplexing method applied to Ethernet switch | |
| CN104407714A (en) | Main board and use method thereof for remote multi-computer switching | |
| CN210270889U (en) | Intelligent analysis expansion equipment and server | |
| CN204759309U (en) | USB, GBE, mixed backplate of PCI -E multibus | |
| CN104516852B (en) | The circuit of I/O links divides multiplexing | |
| CN105824375A (en) | Server | |
| CN217739896U (en) | Intelligent calculation module circuit, calculation board card and computer | |
| CN218547399U (en) | Server based on double-circuit SW3231 processor | |
| CN103440023A (en) | Method and system for service board extension based on ATCA |
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 |