CN109828644B

CN109828644B - Server

Info

Publication number: CN109828644B
Application number: CN201811629106.2A
Authority: CN
Inventors: 余宇
Original assignee: Zhongke Controllable Information Industry Co Ltd
Current assignee: Zhongke Controllable Information Industry Co Ltd
Priority date: 2018-12-28
Filing date: 2018-12-28
Publication date: 2023-04-18
Anticipated expiration: 2038-12-28
Also published as: CN109828644A

Abstract

The invention provides a server (10), wherein the server (10) comprises a first adapter card (12) and a second adapter card (14) which are arranged in the server (10) in parallel and vertically; the first adapter card (12) comprises a first side (122) and a second side (124) which are opposite, the first side (122) is provided with a first notch (126), and the second side (124) is provided with a first clamping groove (128); the second adapter card (14) comprises a first edge (142) and a second edge (144) which are opposite, the first edge (142) is provided with a second notch (146), and the second edge (144) is provided with a second clamping groove (148); and the first card slot (128) passes through the second notch (146), and the second card slot (148) passes through the first notch (126). The invention aims to provide a server for improving the utilization rate of an internal space.

Description

Server

Technical Field

The invention relates to the field of computers, in particular to a server.

Background

4 GPUs are required to be arranged in the existing 2U case, and if the four GPUs are in a sequentially tiled design scheme, the physical space in the case is insufficient.

The existing chassis adopts the upper and lower layer design of 4 GPUs, but because most of GPU cards are designed as double-width cards, a 2U chassis mainboard needs to be cut off for avoiding. However, this design has several disadvantages: PCB cost is wasted; the wiring difficulty is increased because the main board panel becomes smaller; the IO placing space is insufficient, and the IO placing space can only be converted through a high-density interface to meet the requirements of customers, so that the cost is increased, and the users are inconvenient to use.

Disclosure of Invention

In view of the problems in the related art, an object of the present invention is to provide a server for improving utilization of an internal space.

In order to achieve the above object, the present invention provides a server, including a first riser card and a second riser card installed in parallel and vertically in the server; the first adapter card comprises a first side and a second side which are opposite, the first side is provided with a first notch, and the second side is provided with a first clamping groove; the second adapter card comprises a first edge and a second edge which are opposite, the first edge is provided with a second notch, and the second edge is provided with a second clamping groove; and the first clamping groove passes through the second notch, and the second clamping groove passes through the first notch.

According to one embodiment of the invention, the first riser card and the second riser card are spaced apart.

According to one embodiment of the invention, the first riser card and the second riser card are mounted on a bottom surface of the server.

According to one embodiment of the invention, a server card slot is arranged on the bottom surface of the server, and the first adapter card and the second adapter card are inserted into the server card slot.

According to one embodiment of the invention, the bottom surface of the server is provided with an adapter card base, and the first adapter card and the second adapter card are detachably arranged in the adapter card base.

According to one embodiment of the invention, the server further comprises a first GPU board and a second GPU board, wherein the first GPU board is inserted into the first card slot, and the second GPU board is inserted into the second card slot.

According to one embodiment of the present invention, the first GPU board and the second GPU board are disposed in parallel, the first GPU board is disposed vertically to the first riser card and the second riser card, and the first GPU board and the second GPU board are spaced apart in a height direction of the server.

According to one embodiment of the invention, a plurality of groups of first adapter cards and second adapter cards are installed in the server.

The invention has the beneficial technical effects that:

the two adapter cards are arranged in parallel, the clamping grooves of the two adapter cards are mutually inserted, and the clamping grooves of the two adapter cards have certain height difference, so that more spaces capable of containing other parts such as a mainboard are formed inside the server, and the utilization rate of the internal space of the server is improved.

Drawings

FIG. 1 is a schematic diagram of a server of one embodiment of the present invention;

FIG. 2 is a schematic diagram of the position relationship between the adapter card and the GPU board according to one embodiment of the present invention;

FIG. 3 is a schematic view of a first adapter card of one embodiment of the present invention;

FIG. 4 is a schematic diagram of a third riser card according to one embodiment of the invention.

Detailed Description

Embodiments of the server according to the present invention will be described in detail below with reference to the accompanying drawings of fig. 1 to 3.

As shown in fig. 1, in one embodiment of the present invention, the server 10 includes a first riser card 12 and a second riser card 14 mounted side-by-side and vertically in the server 10; the first adapter card 12 includes opposite first and

second sides

122, 124, the first side 122 having a first notch 126, the second side 124 having a first card slot 128; the second riser card 14 includes a first edge 142 and a second edge 144 opposite to each other, the first edge 142 has a second notch 146, and the second edge 144 is provided with a second card slot 148; and first card slot 128 passes through second notch 146 and second card slot 148 passes through first notch 126.

In the present embodiment, the first card slot 128 and the second card slot 148 are positioned substantially parallel and spaced apart. This avoids the first card slot 128 and the second card slot 148 being in the same plane and being disposed along a plane parallel to the bottom surface of the server 10. Thereby avoiding multiple GPU boards being simultaneously laid on the server 10 and eventually going beyond the footprint of the server 10. In addition, the first notch 126 and the second notch 146 are also configured to expose the second card slot 148 and the first card slot 128 in opposite positions, so as to facilitate insertion of the GPU board into the two card slots.

In the present embodiment, as shown in fig. 1 to fig. 2, it should be noted that, in the direction perpendicular to the bottom surface of the server 10, a part of the two GPU boards respectively inserted into the two card slots is overlapped, so as to reduce the length of the vertical projection of the two GPU boards on the bottom surface of the server 10. When a plurality of GPU boards are to be provided in the server 10, the server 10 can accommodate these GPU boards. And the first and

second card slots

128, 148 form a height differential.

As shown in fig. 1-2, in one embodiment of the present invention, the first riser card 12 and the second riser card 14 are spaced apart. Of course, the distance of the spacing may be optimally designed depending on the number of GPU boards that the server 10 needs to accommodate and the size of the service 10.

In this embodiment, the spacing is advantageous to increase the overlapping area of the GPU boards inserted into the first card slot 128 and the second card slot 148, i.e. to increase the overlapping size, so as to reduce the size of the projection of the two GPU boards on the bottom surface of the server 10, so that the GPU boards can be accommodated sufficiently.

As shown in fig. 1, in one embodiment of the present application, a first riser card 12 and a second riser card 14 are mounted on the bottom surface of the server 10. The large area of the bottom surface of the server 10 helps the first riser card 12 and the second riser card 14 to be firmly fixed in the server 10. Of course, in one or more other embodiments, the first riser card 12 and the second riser card 14 may be mounted on a side wall of the server 10. The invention is not limited herein as long as the first riser card 12 and the second riser card 14 can be securely fixed in the server 10.

In one embodiment of the present invention, the bottom surface of the server 10 is provided with server card slots, and the first riser card 12 and the second riser card 14 are inserted into the server card slots. In this embodiment, the two adapter cards are inserted into the server card slot, so that the two adapter cards can be in signal connection with the motherboard of the server 10, which is beneficial to the stability of the overall performance of the server 10.

In one embodiment of the present invention, the bottom surface of the server 10 is provided with a riser card base 20, and the first riser card 12 and the second riser card 14 are detachably mounted in the riser card base 20. In this embodiment, the two adapter cards are inserted into the adapter card base 20, so that the two adapter cards can be in signal connection with the motherboard of the server 10, which is beneficial to the overall performance stability of the server 10. As shown in fig. 3, the first riser card 12 is mounted on a riser card base 20.

As shown in FIG. 2, in one embodiment of the present invention, server 10 also includes a first GPU board 16 and a second GPU board 18, where first GPU board 16 is inserted in first card slot 128 and second GPU board 18 is inserted in second card slot 148.

As shown in fig. 2, in one embodiment of the present invention, the first GPU board 16 and the second GPU board 18 are disposed in parallel, the first GPU board 16 is disposed perpendicular to the first riser card 12 and the second riser card 14, and the first GPU board 16 and the second GPU board 18 are spaced apart in the height direction of the server 10.

In the present embodiment, the first GPU board 16 and the second GPU board 18 are disposed in parallel and spaced apart in the height direction of the server 10, so that a relief space is formed in the height direction of the first GPU board 16 and the second GPU board 18, respectively, thereby creating more space for other server parts such as a motherboard to accommodate.

In one embodiment of the present invention, multiple sets of first riser cards 12 and second riser cards 14 are installed in the server 10. In this embodiment, more space can be formed in which the server components can be accommodated.

The following describes an embodiment of the present invention in detail with reference to fig. 1 to 3 of the drawings.

As shown in fig. 1, the server 10 includes 2 sets of riser cards, one set being a first riser card 12 and a second riser card 14. The first riser card 12 and the second riser card 14 are arranged in parallel and in parallel on the bottom side of the server 10. The first card slot 128 of the first riser card 12 passes through the second card slot 148 of the second riser card 14 and the two riser cards are spaced apart. In addition, first GPU board 16 is inserted in first card slot 128 and second GPU board 18 is inserted in second card slot 148. The first GPU board 16 and the second GPU board 18 are disposed in parallel and spaced apart from each other, and have a height difference.

As shown in fig. 2, the first riser card 12 and the second riser card 14 are arranged in parallel. The first card slot 128 of the first riser card 12 passes through the second card slot 148 of the second riser card 14 and the two riser cards are spaced apart. In addition, first GPU board 16 is inserted in first card slot 128 and second GPU board 18 is inserted in second card slot 148. The first GPU board 16 and the second GPU board 18 are arranged in parallel and spaced apart, and have a height difference. Also, the first GPU board 16 is perpendicular to the first riser card 12 and the second riser card 14, and the second GPU board 18 is perpendicular to the second riser card 14 and the first riser card 12. The first card slot 128 is located on the second side 124 of the first adapter card 12, and the first side 122 of the first card slot 128 is a first notch 126. The second card slot 148 is located on the second side 144 of the second riser card 14, and the first notch 126 is located on the first side 122 of the first card slot 128.

As shown in fig. 3, the first side 122 of the first adapter card 12 is a first notch 126 and the second side 124 is a first card slot 128. The first riser card 12 is mounted in a riser card base 20.

As shown in fig. 4, yet another embodiment of an adapter card, a third adapter card 15, is shown. The first side 151 of the third adapter card 15 is provided with a third notch 155, the second side 153 is provided with a fourth notch 157 larger than the third notch 155, and an edge of the fourth notch 157 is provided with a third slot 159. In one or more embodiments, a third riser card 15 may replace the first riser card 12 or the second riser card 14. Of course, the third riser card 15 may also be installed in the server 10 at the same time as the first and

second riser cards

12, 14. Of course, in other embodiments, two third riser cards 15 are installed in parallel in the server 10, and the third card slot 159 of one third riser card 15 passes through the third notch 155 or the fourth notch 157 of the other third riser card 15.

The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A server (10), characterized in that the server (10) comprises a first riser card (12) and a second riser card (14) mounted side by side and vertically in the server (10);

the first adapter card (12) comprises a first side (122) and a second side (124) which are opposite, the first side (122) is provided with a first notch (126), and the second side (124) is provided with a first clamping groove (128);

the second adapter card (14) comprises a first edge (142) and a second edge (144) which are opposite, the first edge (142) is provided with a second notch (146), and the second edge (144) is provided with a second clamping groove (148); and

the first card slot (128) passing through the second notch (146), the second card slot (148) passing through the first notch (126),

wherein the first side (122) is a top edge of the first adapter card (12), the first notch (126) is a notch that is recessed from the top edge of the first adapter card (12) toward a bottom edge of the first adapter card (12),

the first edge (142) is a bottom edge of the second riser card (14), and the second notch (146) is a notch that is recessed from the bottom edge of the second riser card (14) toward a top edge of the second riser card (14).

2. The server according to claim 1, wherein the first riser card (12) and the second riser card (14) are spaced apart.

3. The server according to claim 1, wherein the first riser card (12) and the second riser card (14) are mounted on a bottom surface of the server (10).

4. The server according to claim 1, characterized in that the bottom surface of the server (10) is provided with a server (10) card slot, and the first adapter card (12) and the second adapter card (14) are inserted in the server (10) card slot.

5. The server according to claim 1, characterized in that the bottom surface of the server (10) is provided with an adapter card base (20), the first adapter card (12) and the second adapter card (14) being detachably mounted in the adapter card base (20).

6. The server according to claim 1, wherein the server (10) further comprises a first GPU board (16) and a second GPU board (18), wherein the first GPU board (16) is inserted in the first card slot (128) and the second GPU board (18) is inserted in the second card slot (148).

7. The server according to claim 6, characterized in that the first GPU board (16) and the second GPU board (18) are arranged in parallel, the first GPU board (16) is arranged perpendicular to the first riser card (12) and the second riser card (14), and the first GPU board (16) and the second GPU board (18) are spaced apart in the height direction of the server (10).

8. The server according to claim 1, characterized in that a plurality of sets of said first riser card (12) and said second riser card (14) are installed in said server (10).