CXL architecture server

Abstract

The embodiment of the utility model discloses a CXL architecture server, which comprises a case, and a storage device and a switch board which are arranged in the case, a CXL chip is arranged in the storage device, the switch board comprises a first connector and a second connector, the first connector is electrically connected with the storage device, the second connector is used for being electrically connected with other servers, and the first connector is electrically connected with the storage device. And the exchange board is used for interconnecting memories of the CXL architecture server and other servers. According to the CXL architecture server provided by the utility model, the CXL chip can be utilized to efficiently interconnect the memory resources of the storage equipment with other servers. According to the method, the memory resources can be shared or expanded across the servers, and the whole server becomes a virtual memory pool, so that the problem of insufficient memory bandwidth and capacity of a single server kernel is solved, and waste caused by idle or insufficient resources is reduced.

Description

Technical Field

[0001] The utility model relates to the field of computer technology, and in particular to a CXL architecture server. Background Art

[0002] The rapid development of AI and big data applications, coupled with the evolution of related technologies, has led to a dramatic increase in the number of server CPU (Central Processing Unit) cores to meet the growing demand for data processing. However, this growth has not led to a simultaneous increase in the memory bandwidth and capacity of each core. Instead, due to resource allocation constraints, the memory bandwidth enjoyed by each core has shown a downward trend. This has directly impacted the overall performance of multi-core CPUs, resulting in underutilization of expensive memory resources and increased data center operating costs. Utility Model Content

[0003] The purpose of the present invention is to overcome the deficiencies of the prior art and provide a CXL architecture server to solve the problem that the expensive memory resources in the existing servers are poorly utilized, thereby increasing the operating costs of the data center.

[0004] In order to achieve the above purpose, the utility model adopts the following technical solutions:

[0005] An embodiment of the present invention provides a CXL architecture server, comprising: a chassis, and a storage device and a switch board built into the chassis. The storage device has a built-in CXL chip. The switch board includes a first connector and a second connector. The first connector is electrically connected to the storage device, and the second connector is used to electrically connect to other servers. The switch board is used to interconnect the memory of the CXL architecture server with the memory of other servers.

[0006] As a preferred technical solution of the present invention, the other server is a server with a built-in RETIMER card, and the second connector is used to electrically connect to the RETIMER card.

[0007] As a preferred technical solution of the present invention, a RETIMER card is further provided in the chassis. The RETIMER card includes a third connector. The third connector is used to electrically connect to the second connector.

[0008] As a preferred technical solution of the present invention, the storage device is a memory hard disk.

[0009] As an optimal technical solution of the present invention, a backplane is also provided in the chassis, the memory hard disk is installed on the backplane, and a fourth connector electrically connected to the memory hard disk is provided on the backplane, and the fourth connector is also electrically connected to the first connector.

[0010] As an optimal technical solution of the present invention, a fan module is further provided in the chassis.

[0011] As a preferred technical solution of the present invention, the memory hard disk is arranged on the front side of the chassis, the switch board is arranged on the rear side of the chassis, and the fan module is arranged on the middle side of the chassis.

[0012] As a preferred technical solution of the present invention, the storage device is a memory expansion card.

[0013] As an optimal technical solution of the present invention, a fan module is further provided in the chassis.

[0014] As an optimal technical solution of the present invention, the fan module is arranged at the front side of the chassis, the switch board is arranged at the rear side of the chassis, and the memory expansion card is arranged at the middle side of the chassis.

[0015] Compared to existing technologies, the CXL architecture server of this utility model utilizes CXL chips to efficiently interconnect the memory resources of storage devices with other servers. This allows memory resources to be shared or expanded across servers, turning the entire server into a virtual memory pool. This alleviates the problem of insufficient kernel memory bandwidth and capacity on a single server and reduces waste caused by idle or insufficient resources.

[0016] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented in accordance with the contents of the specification. In addition, in order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the following preferred embodiments are specifically cited and described in detail as follows. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 This is a three-dimensional view of a CXL architecture server in a first embodiment of the present invention;

[0019] Figure 2 This is a top view of a CXL architecture server in a first embodiment of the present invention;

[0020] Figure 3This is an exploded diagram of a CXL architecture server in a first embodiment of the present invention;

[0021] Figure 4 This is a three-dimensional view of a CXL architecture server in a second embodiment of the present invention;

[0022] Figure 5 This is a top view of a CXL architecture server in a second embodiment of the present invention;

[0023] Figure 6 This is an exploded diagram of a CXL architecture server in a second embodiment of the present invention;

[0024] Description of the symbols in the figure:

[0025] 1. Chassis; 2. Memory and hard disk; 3. Memory expansion card; 4. Switch board; 5. Fan module. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0027] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative efforts shall fall within the scope of protection of the present invention.

[0028] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like indicate orientations or positional relationships based on the orientations or positional relationships described in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0029] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature specified as "first" or "second" may explicitly or implicitly include one or more of such features. In the description of this utility model, "plurality" means two or more, unless otherwise specifically defined.

[0030] In this utility model, unless otherwise specified or limited, the terms "installed," "connected," "connect," "fixed," etc. should be understood in a broad sense. For example, they can refer to connection, detachable connection, or integration; mechanical connection or electrical connection; direct connection or indirect connection through an intermediate medium; internal communication between two components or interaction between two components. For those skilled in the art, the specific meanings of the above terms in this utility model can be understood according to specific circumstances.

[0031] In the present invention, unless otherwise expressly specified or limited, a first feature being "above" or "below" a second feature may include the first and second features being in direct contact, or may include the first and second features being in contact not directly but through another feature between them. Moreover, a first feature being "above," "above," and "above" a second feature may include the first feature being directly above or obliquely above the second feature, or may simply mean that the first feature is higher in level than the second feature. A first feature being "below," "below," and "below" a second feature may include the first feature being directly below or obliquely below the second feature, or may simply mean that the first feature is lower in level than the second feature.

[0032] In the description of this specification, the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic expressions of the above terms should not be understood as necessarily referring to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine the different embodiments or examples described in this specification.

[0033] It should be explained that this solution is designed based on the CXL (Compute Express Link) technology that exists in the existing technology. CXL technology is a new type of high-speed interconnection technology that aims to bridge the memory gap between CPUs and devices, provide higher data throughput and lower latency, and solve the problems of resource waste, inconvenience in operation and performance degradation caused by memory segmentation. This solution specifically integrates CXL technology into server equipment and proposes a new type of CXL architecture server to solve the defects of low server memory resource utilization in the existing technology and increased operating costs of data centers. This is explained in detail below.

[0034] See also Figures 1 to 6, an embodiment of the utility model discloses a CXL architecture server, which includes: a chassis 1, and a storage device and a switch board 4 built into the chassis 1, the storage device has a built-in CXL chip, the switch board 4 includes a first connector and a second connector, the first connector is electrically connected to the storage device, the second connector is used to electrically connect to other servers, and the switch board 4 is used to interconnect the memory of the CXL architecture server with the other servers. Optionally, the first connector in this embodiment is an MCIO connector, and the second connector is a CDFP connector. Optionally, the first connector in this embodiment is electrically connected to the storage device through a high-speed line, and the second connector is electrically connected to other servers through a high-speed line. It should be explained that the CXL chip in this embodiment enables the storage device to support CXL. In this embodiment, the CXL architecture server realizes direct access to memory and high-speed data exchange between servers through high-speed CXL connection, significantly improving the overall performance of the system, allowing memory resources between servers to be flexibly expanded and shared, and meeting the needs of data centers and cloud computing for dynamic resource expansion.

[0035] Furthermore, the other servers are servers with built-in RETIMER cards, and the second connector is used to electrically connect the RETIMER cards. It should be explained that the RETIMER card can be inserted into the signal transmission path to regenerate and resynchronize attenuated or distorted signals to restore the integrity and quality of the signals. When the signals are sent from the switch board 4 of the CXL architecture server and reach other servers through the transmission medium, the RETIMER card will receive these signals and perform necessary signal processing and regeneration. By resynchronizing and amplifying the signals, the RETIMER card can ensure that the signals still maintain high quality and stability when they reach the target server.

[0036] Furthermore, a retimer card is provided in the chassis 1. The retimer card includes a third connector, which is used to electrically connect to the second connector. Specifically, the third connector is used to electrically connect to the second connector of another server. Optionally, the third connector in this embodiment is a CDFP connector.

[0037] like Figure 1-Figure 3 As shown, in one embodiment, the storage device is a memory hard disk 2.

[0038] Furthermore, a backplane is provided within the chassis 1, on which the memory hard disks 2 are mounted. The backplane is provided with a fourth connector electrically connected to the memory hard disks 2, and the fourth connector is also electrically connected to the first connector. Optionally, in this embodiment, a backplane is provided, with 24 memory hard disks 2 mounted on one side and 24 fourth connectors provided on the other side, with the 24 fourth connectors corresponding to the 24 memory hard disks 2 respectively; the switch board 4 is provided with 24 first connectors, with the 24 first connectors corresponding to the 24 fourth connectors respectively. Thus, the memory of all the memory hard disks 2 is transmitted to the first connector via the fourth connectors to the switch board 4, and then the switch board 4 is connected to other servers.

[0039] Furthermore, a fan module 5 is provided within the chassis 1. Optionally, the fan module 5 is a fan wall composed of multiple 8056 fans. Preferably, the fan module 5 is a fan wall composed of four 8056 fans. In this embodiment, the fan module 5 forms an effective heat dissipation duct within the chassis 1, promptly dissipating heat generated within the chassis 1. This protects the server hardware from high-temperature damage, ensures stable operation of the CXL architecture server at an appropriate temperature, extends its service life, and improves the overall stability and reliability of the server.

[0040] Specifically, the layout in this embodiment is that the memory hard disk 2 is arranged at the front side of the chassis 1, the switch board 4 is arranged at the rear side of the chassis 1, and the fan module 5 is arranged at the middle side of the chassis 1.

[0041] like Figure 4-Figure 6 As shown, in another embodiment, the storage device is a memory expansion card 3. Furthermore, the memory expansion card 3 is provided with a fifth connector, which is electrically connected to the first connector. Optionally, the fifth connector is electrically connected to the first connector via a high-speed cable. Optionally, the fifth connector is an MCIO connector.

[0042] Furthermore, a fan module 5 is provided within the chassis 1. Optionally, the fan module 5 is a fan wall composed of multiple 8056 fans. Preferably, the fan module 5 is a fan wall composed of four 8056 fans. In this embodiment, the fan module 5 forms an effective heat dissipation duct within the chassis 1, promptly dissipating heat generated within the chassis 1. This protects the server hardware from high-temperature damage, ensures stable operation of the CXL architecture server at an appropriate temperature, extends its service life, and improves the overall stability and reliability of the server.

[0043] Specifically, the fan module 5 is disposed on the front side of the chassis 1, the switch board 4 is disposed on the rear side of the chassis 1, and the memory expansion card 3 is disposed on the middle side of the chassis 1. Optionally, in this embodiment, eight memory expansion cards 3 are disposed in the chassis 1, and the eight memory expansion cards 3 are stacked in a double layer, with four cards on the upper layer and four cards on the lower layer, on the middle side of the chassis 1. Two switch boards 4 are also disposed in the chassis 1, and the two switch boards 4 are stacked in a double layer, with one card on the upper layer and one card on the lower layer, on the rear side of the chassis 1. Each switch board 4 is provided with four first connectors, and each memory expansion card 3 is provided with one fifth connector. The four memory expansion cards 3 on the upper layer are electrically connected to the switch board 4 on the upper layer via high-speed cables, and the four memory expansion cards 3 on the lower layer are electrically connected to the switch board 4 on the lower layer via high-speed cables. The switch boards 4 are then electrically connected to other servers.

[0044] Compared to existing technologies, the CXL architecture server of this utility model utilizes CXL chips to efficiently interconnect the memory resources of storage devices with other servers. This allows memory resources to be shared or expanded across servers, turning the entire server into a virtual memory pool. This alleviates the problem of insufficient kernel memory bandwidth and capacity on a single server and reduces waste caused by idle or insufficient resources.

[0045] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and such modifications or substitutions are intended to be within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be subject to the scope of protection of the claims.

Claims

1. A CXL architecture server, characterized in that: include: A chassis, a storage device and a switch board built into the chassis, wherein the storage device has a built-in CXL chip, and the switch board includes a first connector and a second connector. The first connector is electrically connected to the storage device, and the second connector is used to electrically connect to other servers. The switch board is used to interconnect the memory of the CXL architecture server with that of other servers.

2. A CXL architecture server according to claim 1, characterized in that: The other server is a server with a built-in RETI MER card, and the second connector is used to electrically connect to the RETI MER card.

3. The CXL architecture server according to claim 1, characterized in that: A RET IMER card is further provided in the chassis. The RET IMER card includes a third connector. The third connector is used to electrically connect to the second connector.

4. The CXL architecture server according to claim 1, characterized in that: The storage device is a memory hard disk.

5. The CXL architecture server according to claim 4, characterized in that: A backplane is further provided in the chassis, the memory hard disk is mounted on the backplane, a fourth connector electrically connected to the memory hard disk is provided on the backplane, and the fourth connector is also electrically connected to the first connector.

6. The CXL architecture server according to claim 4, characterized in that: A fan module is also provided in the chassis.

7. The CXL architecture server according to claim 6, characterized in that: The memory hard disk is arranged at the front side of the chassis, the exchange board is arranged at the rear side of the chassis, and the fan module is arranged at the middle side of the chassis.

8. The CXL architecture server according to claim 7, characterized in that: The storage device is a memory expansion card.

9. The CXL architecture server according to claim 8, characterized in that: A fan module is also provided in the chassis.

10. The CXL architecture server according to claim 8, characterized in that: The fan module is arranged at the front side of the chassis, the switch board is arranged at the rear side of the chassis, and the memory expansion card is arranged at the middle side of the chassis.

Images