CN214011980U

CN214011980U - Server with RAS (remote server system) characteristic

Info

Publication number: CN214011980U
Application number: CN202023037203.8U
Authority: CN
Inventors: 黄凯
Original assignee: Suzhou Inspur Intelligent Technology Co Ltd
Current assignee: Suzhou Inspur Intelligent Technology Co Ltd
Priority date: 2020-12-16
Filing date: 2020-12-16
Publication date: 2021-08-20
Anticipated expiration: 2030-12-16

Abstract

The utility model provides a server with RAS characteristic, include: 4 computing boards, each of which is provided with 2 CPUs; 4 computing boards are interconnected through a UPI bus; each computing board supports 32 memory banks, 2 standard PCIE slots and 2 NVME SSD backplanes; the 4 computing boards are connected with the middle back board through a high-speed connector; the middle backboard is connected with the two GPUs and the two storage expansion boards. The utility model discloses a design system architecture and module interconnection mode, the scalability of memory and CPU allows the scale of system expansion in order to satisfy ever-increasing performance and workload requirement. The CPU expansion is the capacity of increasing the number and functions of CPUs in the system, the system can expand 8 CPUs at most, the memory capacity of the server can be increased by expanding the memory, the system supports 128 memories in total, and the system capacity can meet the change of service requirements.

Description

Server with RAS (remote server system) characteristic

Technical Field

The utility model belongs to the technical field of the server, concretely relates to server with RAS characteristic.

Background

The RAS characteristics are reliability, availability, scalability. The method comprises the steps of large-scale database management and analysis, business application (enterprise resource management, customer relationship management and supply chain management) of an enterprise, virtualization, memory calculation and artificial intelligence development, wherein the core key services have higher and higher requirements on the reliability of a server, in order to bear the key applications, the continuity of the services is ensured, the product value is improved, the RAS characteristic of the product needs to be improved, and under the premise that the outgoing line of a certain related component fails, the corresponding component supports thermal maintenance, so that the customer services can be more effectively prevented from being influenced.

The client has higher requirement on storage expansion of PCIE equipment according to service requirements, if the number of the supported PCIE equipment is less, the configuration is single, and the flexibility is not high, the application range of the server is directly influenced, the same platform product meets various application requirements of the client, and compared with a special machine, the PCIE equipment has stronger competitiveness and is convenient for flexible expansion during service upgrade.

SUMMERY OF THE UTILITY MODEL

To the above insufficiency of the prior art, the present invention provides a server having RAS characteristics to solve the above technical problem.

The utility model provides a server with RAS characteristic, include:

4 computing boards, each of which is provided with 2 CPUs; 4 computing boards are interconnected through a UPI bus; each computing board supports 32 memory banks, 2 standard PCIE slots and 2 NVME SSD backplanes; the 4 computing boards are connected with the middle back board through a high-speed connector; the middle backboard is connected with the two GPUs and the two storage expansion boards.

Further, the south bridge of the server adopts a buckle structure.

Further, the upper layer of the middle backboard is connected with the first GPU and the first storage expansion board; the lower layer of the middle backboard is connected with a second GPU and a second storage expansion board; the first GPU, the first storage expansion board, the second GPU and the second storage expansion board correspond to 6 fans; the fan is connected with the middle back plate through the connector.

Furthermore, a plurality of PCIE slots are arranged on the middle backboard and are connected with the GPU through the PCIE slots; and the middle back plate is also provided with a storage expansion plate, and the storage expansion plate adopts a RAID card.

Further, the server comprises 6 standard redundant power supplies, four network cards, 8 system serial ports, 8 USB2.0 interfaces, 4 RJ45 interfaces and 4 VGA interfaces.

The utility model has the advantages that,

the utility model provides a server with RAS characteristic, through design system architecture and module interconnection mode, the scalability of memory and CPU allows the scale of system expansion in order to satisfy ever-increasing performance and workload requirement. The CPU expansion is the capacity of increasing the number and functions of CPUs in the system, the system can expand 8 CPUs at most, the memory capacity of the server can be increased by expanding the memory, the system supports 128 memories in total, and the system capacity can meet the change of service requirements.

Furthermore, the utility model relates to a principle is reliable, and simple structure has very extensive application prospect.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.

FIG. 1 is a schematic diagram of a front window of a server according to an embodiment of the present application;

FIG. 2 is a schematic illustration of a rear window of a server according to an embodiment of the present application;

FIG. 3 is a schematic diagram illustrating a side view of an architectural interconnection of servers according to one embodiment of the present application;

FIG. 4 is a schematic top view of a server according to one embodiment of the present application;

FIG. 5 is a diagram of a GPU and a memory expansion board of a server according to an embodiment of the present application.

Detailed Description

In order to make the technical solutions in the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only some embodiments of the present invention, but not all embodiments. Based on the embodiments in the present invention, all other embodiments obtained by a person skilled in the art without creative efforts shall belong to the protection scope of the present invention.

It should be noted that, in the present invention, the embodiments and features of the embodiments may be combined with each other without conflict.

In the description of the present invention, it is to be understood that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", and the like, indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are used merely for convenience of description and for simplicity of description, and do not indicate or imply that the device or element being referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore, should not be construed as limiting the present invention. Furthermore, the terms "first", "second", etc. are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first," "second," etc. may explicitly or implicitly include one or more of that feature. In the description of the present invention, "a plurality" means two or more unless otherwise specified.

In the description of the present invention, it is to be noted that, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meaning of the above terms in the present invention can be understood by those of ordinary skill in the art through specific situations.

The present invention will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.

Example 1

The embodiment provides a server with RAS characteristics, and based on an Intel platform and a latest processor, a novel 8-way server is developed, a computing board, a GPU (graphics processing Unit) and a storage expansion board are designed and researched, a middle backboard, a fan board and a hard disk backboard, wherein the computing board, the GPU and the storage expansion board and the fan board are connected together through the middle backboard, and the hard disk backboard is connected to the GPU and the storage expansion board through a cable. The chassis is 8U, and the whole size of system is 730.0 447.0 351.0 mm. The concrete structure is as follows:

as shown in fig. 1, the front window of the chassis includes: 2 VGA interfaces and 4 USB2.0 interfaces; 4 GPUs; 8 PCIE slots; support 24 hard disks (2 12-disk backplane assemblies), 8 NVME SSD.

As shown in fig. 2, the rear window of the chassis includes: 12 PCIE slots supporting hot plugging (wherein 6 PCIE risers are used to implement expansion, and PCIE modules all support hot plugging); 6 standard redundant power supplies supporting N + N redundancy; 8 fans, which support N +1 redundancy and can support thermal maintenance; 4 OCP3.0 network cards (where 2 OCP riser boards are used to implement expansion, each OCP riser board can support 2 OCP3.0 network cards, and support 4 OCP3.0 in total, where a PCIE signal of each OCP3.0 comes from one computing board to implement balanced design); system serial port 8, USB2.0 interface 8, RJ45 interface 4, VGA interface 4.

The connection relationship among the board cards is shown in fig. 3, 4 computing boards are connected to the middle back board through high-speed connectors, and CPUs on different computing boards are interconnected through a UPI bus; the upper part and the lower part of the case are respectively provided with 1 GPU and a storage expansion board, the storage expansion boards are connected to the middle back board through high-speed connectors and used for expanding the GPUs and the hard disk, PCIE slots are designed on the board cards, and the PCIE expansion boards are expanded and used for installing the GPUs; the RAID card (storage expansion board) is designed on the board card and used for signal input of the hard disk backboard, PCIE signals of the PCIE slot and the RAID card are from a CPU of the computing node, and the PCIE signals of the CPU are connected to the GPU and the storage expansion board through the middle backboard; each GPU and each storage expansion board are provided with 6 fans for dissipating heat of the GPU and the hard disk.

The PCIE IO board is connected to the middle back board through a high-speed connector and used for expanding the PCIE equipment; the fan board is connected to the middle back board through the connector, and the fan module is expanded to dissipate heat for the 4 computing nodes.

As shown in fig. 4, each computing board has 2 CPUs, and 4 computing boards have 8 CPUs; a single computing board can support 32 DDR4, and 4 computing boards support 128 memory banks; the 4 computing boards are connected to the middle back board through high-speed connectors, and each computing board is interconnected with the other 3 computing boards through a high-speed bus UPI; the (PCH) south bridge is designed in a buckle form; a single computing board supports 2 standard PCIE slots, and 4 computing boards support 8 standard PCIE slots in total and are used for expanding PCIE equipment; the single computing board supports 2 NVME SSDs, wherein PCIE signals of a back board of the NVME SSDs come from a CPU; the high-speed bus PCIE input by the CPU is connected to the middle back plate and then goes to PCIE slots and RAID cards on the GPU and the storage expansion plate, and is used for supporting the GPU of the whole system and expanding the storage hard disk back plate.

As shown in fig. 5, the GPU and the storage expansion board are connected to the midplane via a high-speed connector, and can support 2 GPUs and 12 hard disks, and the system supports 4 GPUs and 24 hard disks; the 2 GPUs are expanded through a PCIE expansion board, PCIE signals of the PCIE expansion board are output from a CPU of the computing board, and interconnection is achieved through the middle backboard; 12 hard disks are installed on a hard disk backboard, the hard disk backboard is connected to an RAID buckle card in an uplink mode, PCIE signals of the RAID buckle card are output from a CPU of a computing board, and interconnection is achieved through a middle backboard; USB, VGA expansion board can provide USB interface and VGA interface of the front window of the server; the fan modules on the GPU and the storage expansion board dissipate heat for the GPU and the hard disk, wherein the fan modules are directly installed on the GPU and the storage expansion board, and the rotating speed of the fan is controlled by monitoring and regulating the speed of the BMC on the computing board.

Customer traffic demands for workloads are growing, requiring high computing power and configuration scalability to support, and system architectural design and modular interconnect approaches, such as memory and CPU scalability, in the solution allow systems to expand to meet the growing size of performance and workload demands. The CPU expansion is the capacity of increasing the number and functions of CPUs in the system, the system can expand 8 CPUs at most, the memory capacity of the server can be increased by expanding the memory, the system supports 128 memories in total, and the system capacity can meet the change of service requirements.

And the system-level components are designed in a redundant manner, such as fan redundancy, power supply redundancy, storage redundancy and the like, so that the reliability of the system is guaranteed.

The 4 computing boards can realize partition management through the interconnection framework of the middle back panel, carry out quick partition according to the requirements of customers, and can be used as 1 eight-path server, 2 four-path servers and 4 two-path servers;

high-performance computing has a plurality of computing tasks which are operated in parallel, high-performance high-end servers cannot be expanded, and working time of 4 computing boards can be guaranteed by the scheme;

according to the scheme, 4 computing boards, 20 PCIE equipment expansion, 4 GPUs and 32 storage hard disks are highly integrated, higher hardware resource utilization is achieved, only one physical machine simplifies basic framework management, redundancy of a plurality of modules guarantees continuity of core services, and data safety of core application is guaranteed.

Although the present invention has been described in detail by referring to the drawings in conjunction with the preferred embodiments, the present invention is not limited thereto. Various equivalent modifications or substitutions can be made on the embodiments of the present invention by those skilled in the art without departing from the spirit and substance of the present invention, and these modifications or substitutions are intended to be within the scope of the present invention/any person skilled in the art can easily conceive of changes or substitutions within the technical scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A server having RAS characteristics, comprising:

2. The server according to claim 1, wherein the south bridge of the server employs a card-in-card architecture.

3. The server according to claim 1, wherein the mid-backplane upper layer connects the first GPU and the first storage expansion board; the lower layer of the middle backboard is connected with a second GPU and a second storage expansion board; the first GPU, the first storage expansion board, the second GPU and the second storage expansion board correspond to 6 fans; the fan is connected with the middle back plate through the connector.

4. The server according to claim 1, wherein the middle backplane is provided with a plurality of PCIE slots, and the PCIE slots are connected to the GPU; and the middle back plate is also provided with a storage expansion plate, and the storage expansion plate adopts a RAID card.

5. The server according to claim 1, wherein the server comprises 6 standard redundant power supplies, four network cards, 8 system serial ports, 8 USB2.0 interfaces, 4 RJ45 interfaces, and 4 VGA interfaces.