Hard disk backplane and server

Abstract

The application provides a hard disk backboard and a server, the hard disk backboard is provided with a plurality of first connectors and a plurality of second connectors, and the hard disk backboard can be switched between a first working mode and a second working mode; in the first working mode, each first connector and each second connector are respectively used for electrically connecting different hard disks, and each hard disk is in communication connection with a mainboard through the first connector or the second connector; in the second working mode, at least two first connectors are matched to connect a RAID chip, and the other first connectors or the second connectors are respectively used for electrically connecting different hard disks, the RAID chip is in communication connection with the corresponding hard disk through the corresponding first connector, and the RAID chip is in communication connection with the mainboard through the first connector. The hard disk backboard and the server provided by the application not only improve the expansibility and flexibility of the system, but also realize efficient management and configuration of the hard disks, and play a role in fault isolation, hard disk redundancy and performance stability.

Description

Technical Field

[0001] This invention relates to the field of server equipment technology, and more particularly to a hard disk backplane and a server. Background Technology

[0002] With the rapid development of data centers and cloud computing technologies, server systems are increasingly demanding higher requirements for data storage capacity, speed, and reliability. Against this backdrop, RAID (Redundant Array of Independent Disks) technology has become a key component of server storage solutions due to its dual advantages of providing data redundancy and performance enhancement.

[0003] Currently, traditional RAID chips are typically deployed at the rear of the server chassis, connected to the motherboard via PCIe or OCP interfaces, and then connected to the hard drive backplane via long cables. This design not only increases system complexity and potential points of failure but also limits the optimal utilization of internal server space. Conversely, directly mounting the RAID chip on the hard drive backplane is limited by fixed interfaces and configuration options, making it difficult to meet rapidly changing market demands. Currently, there is a lack of mature hard drive backplane design solutions on the market that support multiple operating modes. Summary of the Invention

[0004] This invention provides a hard drive backplane and server, which provides a hard drive backplane design scheme that supports multiple working modes, and solves the problem that the method of directly setting the RAID chip on the hard drive backplane is limited to fixed interfaces and configuration options, making it difficult to meet the rapidly changing market demands.

[0005] In a first aspect, the present invention provides a hard disk backplane, wherein the hard disk backplane is provided with a plurality of first connectors and a plurality of second connectors, and the hard disk backplane is capable of switching between a first working mode and a second working mode. In the first working mode, each of the first connectors and each of the second connectors are used to electrically connect different hard drives, and each hard drive is communicatively connected to the motherboard through the first connector or the second connector. In the second operating mode, at least two of the first connectors cooperate to connect to the RAID chip, and the other first connectors or second connectors are used to electrically connect to different hard drives. The RAID chip communicates with the corresponding hard drive through the corresponding first connector, and the RAID chip communicates with the motherboard through the first connector. The RAID chip is configured to manage each hard drive according to the hard drive information read from each hard drive.

[0006] According to the present invention, a hard disk backplane is provided, wherein the first connector is provided with a first interface and the second connector is provided with a second interface; In the first working mode, both the first interface and the second interface are used to connect to the hard disk; In the second operating mode, the two first interfaces are connected to the RAID chip, and the other first interfaces and all the second interfaces are used to electrically connect to different hard drives.

[0007] According to a hard disk backplane provided by the present invention, the first interface is provided with a first pin and a second pin, and the second interface is provided with a second pin; In the first operating mode, the second pin in the first interface and the second pin in the second interface are both used to connect to the hard disk; In the second operating mode, the first pin and the second pin of the two first interfaces are both used to connect to the RAID chip, and the second pin of the other first interface and the second pin of the second interface are both used to connect to the hard drive.

[0008] According to a hard disk backplane provided by the present invention, the RAID chip, in the second working mode, is used to obtain the hard disk information of a plurality of hard disks connected to the hard disk backplane, receive the RAID function enable command, select a target hard disk from the plurality of hard disks according to a preset RAID level and the hard disk information of the plurality of hard disks, and establish an independent hard disk redundancy array for the target hard disk.

[0009] According to a hard drive backplane provided by the present invention, the hard drive backplane is provided with a motherboard connector, the motherboard is connected to the motherboard connector, and the RAID chip is electrically connected to the motherboard connector through both the first connector and the second connector.

[0010] According to the present invention, a hard disk backplane is provided with a plurality of backplanes, each of the backplanes being provided with a first connector and a plurality of second connectors, and each of the backplanes of the hard disk backplane is capable of switching between a first working mode and a second working mode; When any of the backplanes is in the first operating mode, the first connector and each of the second connectors in each of the backplanes are respectively used to electrically connect different hard drives; When any of the backplanes is in the second operating mode, at least two of the first connectors in the backplane are connected to the RAID chip. The other first connectors or second connectors in the backplane are used to electrically connect to different hard drives. The RAID chip in the backplane is connected to the corresponding hard drive through the corresponding first connector. The RAID chip in the backplane is connected to the motherboard through the corresponding first connector. The RAID chip is configured to manage each hard drive in the backplane according to the hard drive information read from each hard drive in the backplane.

[0011] According to the present invention, a hard disk backplane has a plurality of first connectors and a plurality of second connectors arranged in an array.

[0012] According to the present invention, a hard disk backplane further includes: A switching circuit is configured to control the hard disk backplane to switch between a first operating mode and a second operating mode.

[0013] According to the present invention, a hard disk backplane further includes: A switching switch, electrically connected to the switching circuit, is integrated on the hard disk backplane and configured to receive user input or system commands to trigger the switching of working modes. An indicator light, electrically connected to the switching circuit, is used to display the working mode status of the hard drive backplane.

[0014] In a second aspect, the present invention provides a server, comprising: Hard drive backplate; A RAID chip and a motherboard, wherein the RAID chip is communicatively connected to the motherboard via the first connector.

[0015] The hard drive backplane and server provided by this invention can flexibly switch between a first operating mode (direct communication mode) and a second operating mode (RAID chip management mode). In the first operating mode, the first connector and multiple second connectors on the hard drive backplane are directly connected to the hard drive, enabling direct communication between the hard drive and the motherboard. In the second operating mode, two first connectors cooperate to connect to the RAID chip, which then communicates with the motherboard through the connected hard drive and the first connector, achieving efficient management and configuration of the hard drive. This isolates the hard drive from the motherboard, providing fault isolation, hard drive redundancy, and performance stability. The design of the hard drive backplane allows users to flexibly insert RAID chips or hard drives according to actual needs without having to consider complex issues such as cable connections. This plug-and-play design greatly improves the scalability and flexibility of the system, meeting the storage needs of different application scenarios.

[0016] Meanwhile, by integrating the RAID chip's connectivity onto the hard drive backplane, the slot space and location occupied by traditional RAID chips are significantly saved. Furthermore, integrating the RAID chip's connectivity onto the hard drive backplane reduces the need for external RAID chips, thereby improving the RAID chip's heat dissipation. For RAID chips with increasingly higher performance and power consumption, this design helps optimize heat dissipation, ensuring the RAID chip maintains a stable temperature while operating efficiently, extending its lifespan. Eliminating the multiple cables between the traditional RAID chip and the hard drive backplane avoids signal attenuation and link problems caused by excessive cable length or improper layout. This improvement not only ensures signal transmission quality but also reduces costs associated with using numerous cables. Moreover, the simpler cable layout helps optimize internal airflow, improving overall heat dissipation. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of a hard disk backplane in a first operating mode according to an embodiment of the present invention.

[0019] Figure 2 This is a schematic diagram of a hard disk backplane in a second operating mode according to an embodiment of the present invention.

[0020] Figure 3 This is one of the structural schematic diagrams of a hard disk backplane according to an embodiment of the present invention.

[0021] Figure 4 This is a schematic diagram of a RAID chip being inserted into a hard drive backplane according to one embodiment of the present invention; Figure 5 This is a schematic diagram of a RAID chip inserted into a hard drive backplane according to another embodiment of the present invention; Figure 6 This is a schematic diagram of a RAID chip inserted into a hard drive backplane according to one embodiment of the present invention.

[0022] Figure 7 This is a second schematic diagram of the hard disk backplane according to one embodiment of the present invention.

[0023] Figure 8 This is the third schematic diagram of the hard disk backplane according to one embodiment of the present invention.

[0024] Figure label: 100, First connector; 110, First interface; 200, Second connector; 210, Second interface; 300, Hard drive; 400, Motherboard; 500, RAID chip; 600, Backplane. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0026] In current server backplane designs, RAID chips are typically located at the rear of the hard drive backplane, resulting in poor heat dissipation. Airflow passing over the front hard drives and motherboard generates significant heat, impacting the high-power RAID chips. The RAID chips are connected to the backplane via cables, and in some more complex models, these cables can exceed 1800mm in length. This can lead to signal quality degradation and the risk of link failures. Furthermore, the complex cabling can cause cable bends and other issues that could result in link failures. Additionally, this type of hard drive backplane has limited scalability, requiring modifications to the cabling for upgrades.

[0027] To address the above problems, one embodiment of the present invention provides, such as... Figures 1 to 3 As shown, a hard drive backplane is provided, which has multiple first connectors 100 and multiple second connectors 200. The hard drive backplane can switch between a first operating mode and a second operating mode. In the first operating mode, each first connector 100 and each second connector 200 is used to electrically connect to different hard drives 300, and each hard drive 300 is communicatively connected to the motherboard 400 through the first connector 100 or the second connector 200. In the second operating mode, at least two first connectors 100 cooperate to jointly connect to a RAID chip 500, and the other first connectors 100 and second connectors 200 are used to electrically connect to different hard drives 300. The RAID chip 500 is communicatively connected to the corresponding hard drive 300 through the corresponding first connector 100, and the RAID chip 500 is communicatively connected to the motherboard 400 through the first connector 100. The RAID chip 500 is configured to manage each hard drive 300 according to the hard drive 300 information read from it, so as to isolate the hard drives 300 from the motherboard 400, thereby achieving fault isolation, hard drive 300 redundancy, and performance stability.

[0028] Specifically, such as Figure 1 As shown, when the hard drive backplane switches to the first operating mode, in this mode, all the first connectors 100 and all the second connectors 200 on the hard drive backplane are independently used to connect different hard drives 300. Each hard drive 300 communicates directly with the motherboard 400 through its corresponding connector (first connector 100 or second connector 200) to realize data reading, writing, and status feedback. This mode is suitable for application scenarios that do not require complex data redundancy or advanced fault isolation functions, such as single hard drive 300 storage systems or simple data backup systems.

[0029] When the hard drive backplane switches to the second operating mode, such as Figure 2 As shown, in this mode, the operation of the hard drive backplane changes significantly. The first connector 100 no longer connects directly to the hard drive 300, but instead connects to the RAID chip 500. The RAID chip 500 is a device specifically designed for data redundancy and fault recovery, capable of managing multiple hard drives 300 to improve data security. At least two first connectors 100 are connected to the RAID chip 500, enabling the RAID chip 500 to communicate directly with some of the hard drives 300, managing their data read / write operations and fault detection. The remaining first connectors 100 and second connectors 200 continue to connect other hard drives 300, but the data processing and management of these hard drives 300 are all handled by the RAID chip 500. The RAID chip 500 communicates with the motherboard 400 through the first connectors 100, reporting the status and data of all connected hard drives 300 to the motherboard 400, and also receiving instructions from the motherboard 400. The use of the RAID chip 500 isolates faulty hard drives 300, preventing them from affecting the operation of the entire storage system, thereby improving system stability and reliability. The RAID 500 chip enables redundant data storage, ensuring data integrity and accessibility even if some hard drives (300) fail. The RAID 500 chip optimizes data read / write operations, improving the overall performance of the storage system, especially when handling a large number of concurrent read / write requests. This mode is particularly suitable for applications requiring high data security, fault isolation, and redundant storage, such as servers and data centers.

[0030] The hard drive backplane provided in this embodiment of the invention can flexibly switch between a first working mode (direct communication mode) and a second working mode (RAID chip 500 management mode). In the first working mode, the first connector 100 and multiple second connectors 200 on the hard drive backplane are directly connected to the hard drive 300, enabling direct communication between the hard drive 300 and the motherboard 400. In the second working mode, two first connectors 100 cooperate to connect to the RAID chip 500. The RAID chip 500 then communicates with the motherboard 400 through the connected hard drive 300 and the first connectors 100, achieving efficient management and configuration of the hard drive 300. This isolates the hard drive 300 from the motherboard 400, providing fault isolation, redundancy, and performance stability for the hard drive 300. The design of the hard drive backplane allows users to flexibly insert the RAID chip 500 or the hard drive 300 according to actual needs without having to consider complex issues such as cable connections. This plug-and-play design greatly improves the scalability and flexibility of the system, meeting the storage needs of different application scenarios.

[0031] Meanwhile, by integrating the RAID 500 connectivity onto the hard drive backplane, the slot space and location occupied by the traditional RAID 500 chip are significantly reduced. Integrating the RAID 500 connectivity onto the hard drive backplane also reduces the need for an external RAID 500 chip, thereby improving its heat dissipation. For RAID 500 chips with increasingly higher performance and power consumption, this design helps optimize heat dissipation, ensuring stable temperature operation and extending lifespan. Eliminating the multiple cables between the traditional RAID 500 chip and the hard drive backplane avoids signal attenuation and link problems caused by excessive cable length or improper cable layout. This improvement not only ensures signal transmission quality but also reduces costs associated with using numerous cables. Furthermore, the cleaner cable layout helps optimize internal airflow, improving overall heat dissipation.

[0032] It should be noted that, depending on the needs, one or more RAID chips 500 can be installed on a single hard drive backplane. Correspondingly, more first connectors 100 will be required. For example, if the system has high performance requirements, more RAID chips 500 may be needed to distribute the processing load, thereby improving overall performance. Installing multiple RAID chips 500 can facilitate more complex RAID configurations, thus providing greater data security.

[0033] In some embodiments, such as Figure 4 and Figure 5As shown, the first connector 100 has a first interface 110, and the second connector 200 has a second interface 210. In the first operating mode, each first interface 110 and each second interface 210 is used to connect to one hard drive 300. In the second operating mode, both first interfaces 110 are connected to the RAID chip 500, and the other first interfaces 110 and all second interfaces 210 are used to electrically connect to different hard drives 300.

[0034] Specifically, when the hard drive backplane is in the first operating mode, the first interface 110 of each first connector 100 and the second interface 210 of each second connector 200 are directly connected to the hard drive 300. Each interface is responsible for data transmission and power supply (if the interface supports it) to the corresponding hard drive 300. This mode is suitable for scenarios that require maximizing the number of hard drives 300 and storage capacity.

[0035] When the hard drive backplane is in the second operating mode, both first interfaces 110 are connected to the RAID chip 500. The other first interfaces 110 and all second interfaces 210 are used to electrically connect different hard drives 300. The RAID chip 500 is responsible for combining multiple hard drives 300 into a single logical unit, providing data redundancy, increasing data transfer rates, or both. This mode is suitable for scenarios requiring high data availability, fault tolerance, or higher performance, such as enterprise servers and high-performance computing clusters. Through the RAID chip 500, data reliability and access speed can be improved, while reducing the risk of data loss due to the failure of a single hard drive 300.

[0036] In this embodiment, the hard drive backplane can flexibly adjust its connection method according to different working requirements, thereby providing greater flexibility and adaptability. Whether it's the direct hard drive 300 connection mode that requires maximizing storage capacity, or the RAID chip 500 connection mode that requires data redundancy and performance improvement, both can be achieved through this design.

[0037] It should be noted that, depending on the specifications used by the RAID chip 500, the RAID chip 500 can be connected to at least two first interfaces 110.

[0038] For example, such as Figure 4 As shown, the RAID chip 500 is plugged into two first interfaces 110. This configuration allows the RAID chip 500 to communicate simultaneously with the motherboard 400 (via the first interface 110) and the hard drive 300 (via the first interface 110), making it suitable for simple RAID configurations.

[0039] Or, such as Figure 5As shown, the RAID chip 500 is plugged into three first interfaces 110. This configuration provides greater flexibility, allowing the RAID chip 500 to manage more hard drives 300 simultaneously. This is beneficial for complex RAID configurations that require higher data redundancy or higher performance.

[0040] Therefore, when designing a hard drive backplane, it is necessary to select the appropriate interface connection method based on the specifications of the RAID 500 chip, the required RAID configuration, and the overall performance requirements of the system.

[0041] In some embodiments, the first interface 110 is provided with a first pin and a second pin, and the second interface 210 is provided with a second pin; in a first working mode, the second pin in the first interface 110 and the second pin in the second interface 210 are both used to connect to the hard disk 300; in a second working mode, the first pin and the second pin in both first interfaces 110 are both used to connect to the RAID chip 500, and the second pin in the other first interface 110 and the second pin in the second interface 210 are both used to connect to the hard disk 300.

[0042] Specifically, the first interface 110 differs from the second interface 210. The first interface 110 has a first pin and a second pin, while the second interface 210 has a second pin. The first pin and the second pin have different functions, enabling seamless switching between different operating modes. In the first operating mode, the second pin of all interfaces is used to connect to the hard drive 300, ensuring stable communication between the hard drive 300 and the motherboard 400. In the second operating mode, specific first interfaces 110 (such as the two interfaces connecting to the RAID chip 500) use both the first pin and the second pin simultaneously to support the complex communication needs between the RAID chip 500, the hard drive 300, and the motherboard 400, while other interfaces maintain the second pin connection to the hard drive 300, ensuring support for additional hard drives 300. This design not only simplifies hardware connections but also reduces compatibility issues that may arise from changing operating modes, providing users with a more convenient and reliable user experience.

[0043] Furthermore, as technology continues to develop, the interface types and pin configurations of the hard drive 300 and RAID chip 500 may also change. Therefore, the specific configurations of the first interface 110 and the second interface 210 can also be adapted to accommodate different models and specifications of hard drives 300 and RAID chips 500. For example, additional interfaces or pins can be reserved, or programmable logic circuits can be designed on the control circuit board to address potential future interface changes or upgrade requirements.

[0044] In some embodiments, when the RAID chip 500 is in the second working mode, the RAID chip 500 is used to obtain the hard disk 300 information of the multiple hard disks 300 connected to the hard disk backplane, receive the RAID function enable instruction, select the target hard disk 300 from the multiple hard disks 300 according to the preset RAID level and the hard disk 300 information of the multiple hard disks 300, and establish an independent hard disk 300 redundant array for the target hard disk 300.

[0045] Specifically, when the RAID chip 500 is in its second operating mode, its primary task is to acquire detailed information about all the hard drives 300 connected to its backplane. This information includes, but is not limited to, the capacity, model, serial number, health status (such as SMART data), and interface type of each hard drive 300. After collecting information from all hard drives 300, the RAID chip 500 awaits a RAID enable command from the system or administrator. This command instructs the RAID chip 500 to begin configuring RAID based on the preset RAID level and the currently available hard drive 300 resources. The preset RAID level determines how data is distributed and redundant across multiple hard drives 300. For example, RAID 0 provides data striping to improve performance but does not provide data redundancy; RAID 1 provides mirroring redundancy, ensuring high data availability but sacrificing some storage capacity; RAID 5 combines striping and distributed parity, providing good performance and some redundancy. Based on the preset RAID level and the collected hard drive 300 information, the RAID chip 500 selects the most suitable combination of hard drives 300 as the target hard drive 300. Once the target hard drives 300 are identified, the RAID chip 500 begins configuring these hard drives 300 to form a RAID array. This process includes initializing the hard drives 300, allocating storage space, configuring data distribution and redundancy strategies, etc. For example, in a RAID 5 configuration, the RAID chip 500 configures data striping across all hard drives 300 and stores parity information on one of the hard drives 300 to ensure data recovery should any hard drive 300 fail. Once the RAID array is successfully established and initialized, the RAID chip 500 notifies the system or administrator that the RAID configuration has been successfully completed.

[0046] In one example, such as Figures 4 to 6 As shown, the RAID chip 500 uses a 4C or 4C+ interface and can be directly plugged into the first interface 110 reserved on the back panel of the hard drive. When the RAID chip 500 is not used, the first interface 110 can be compatible with the hard drive 300 with an E3.S interface.

[0047] Specifically, the first interface 110 is an SFF-TA-1002 4C interface. SFF-TA-1002 is a standard for small form factor (SFM) telecom adapters, typically used for high-speed data transmission in server and storage systems. The 4C interface means that the connector supports four channels or connection points, a design that allows for higher data transmission rates and more complex signal transmission. Figure 6 As shown, the SFF-TA-1002 4C interface is used as the interface for the RAID chip 500, and its rightmost side is the E3.S interface, which provides sufficient bandwidth and flexibility to support RAID operations.

[0048] The second interface 210 is an E3.S NVMe interface. E3.S is an NVMe (Non-Volatile Memory Express) interface standard for enterprise-class storage. It is based on the M.2 form factor but extended to support higher performance and data transfer rates. NVMe is a PCIe (Peripheral Component Interconnect Express) based storage access and transfer protocol designed specifically for solid-state drives (SSDs) to provide lower latency and higher throughput than traditional SATA SSDs. Here, the E3.S NVMe interface is used as the second interface 210, allowing direct connection of an NVMe-enabled SSD to the drive backplane.

[0049] Because the SFF-TA-1002 4C interface includes an E3.S NVMe interface, this means that, at the physical level, a hard drive 300 with an E3.S NVMe interface can be inserted into the first interface 110 reserved for the RAID chip 500. This greatly improves the system's scalability and flexibility, meeting the storage needs of different application scenarios.

[0050] In some embodiments, such as Figure 4 As shown, the two first interfaces 110 form a RAID chip 500 connector for the insertion of the RAID chip 500 of the 16i.

[0051] Typically, the 16i RAID 500 uses a 2x1TB EDSFF (Enterprise & Data Center Solid State Form Factor) E3 form factor, a solid-state drive (SSD) standard designed specifically for data center and enterprise applications, with specific dimensions and interface specifications. The 2x1TB EDSFF E3 device thickness means that the RAID 500, or chip, was designed to occupy two E3.S slots. This design helps ensure sufficient physical space and heat dissipation to support high-performance RAID operation.

[0052] The first interface 110, located at the bottom, is used to connect to the motherboard 400 (Host). The first interface 110 is equivalent to the PCIe (Peripheral Component Interconnect Express) interface, responsible for connecting the RAID chip 500 to the main processing unit of the computer system (such as the CPU), enabling data transmission and control signal exchange. This is the critical communication path between the RAID chip 500 and the motherboard 400, and is essential for ensuring the normal operation of the RAID chip 500 and data transmission.

[0053] The second interface 210, located at the top, is for connecting end devices. It is used to connect the hard drive 300 or other storage devices to the hard drive backplane. It can be a direct connection to the hard drive backplane or an indirect connection via an expander backplane to enable more complex storage architectures. By supporting direct or expander backplane connections, this design provides greater flexibility and scalability, allowing the RAID chip 500 to adapt to different storage needs and configurations.

[0054] In the first operating mode, the E3.S NVMe interface (or other compatible interface) of each hard drive 300 can be directly inserted into the first interface 110 and the second interface 210 on the hard drive backplane. In the second operating mode, the 4C or 4C+ interface of the RAID chip 500 is inserted into the RAID chip 500 connector on the hard drive backplane. This connector is composed of multiple first interfaces 110.

[0055] In some embodiments, such as Figure 5 As shown, the three first interfaces 110 form a RAID chip 500 connector for the insertion of the RAID chip 500 of the 32i.

[0056] Typically, the 32i RAID 500 uses a 3x1T EDSFF (Enterprise & Data Center Solid State Form Factor) E3 form factor, a solid-state drive (SSD) standard designed specifically for data center and enterprise applications, with specific dimensions and interface specifications. The 3x1T EDSFF E3 device thickness indicates that the RAID 500, or chip, was designed to occupy three E3.S slots. This design allows for complex configurations.

[0057] The first interface 110, located at the bottom, is used to connect to the motherboard 400 (Host). The first interface 110 is equivalent to a PCIe interface, responsible for connecting the RAID chip 500 to the main processing unit of the computer system (such as the CPU), enabling data transmission and control signal exchange. This is the critical communication path between the RAID chip 500 and the motherboard 400, and is essential for ensuring the normal operation of the RAID chip 500 and data transmission.

[0058] The second interface 210, located at the top, is for connecting end devices. It is used to connect the hard drive 300 or other storage devices to the hard drive backplane. It can be a direct connection to the hard drive backplane or an indirect connection via an expander backplane to enable more complex storage architectures. By supporting direct or expander backplane connections, this design provides greater flexibility and scalability, allowing the RAID chip 500 to adapt to different storage needs and configurations.

[0059] In the first operating mode, the E3.S NVMe interface (or other compatible interface) of each hard drive 300 can be directly inserted into the first interface 110 and the second interface 210 on the hard drive backplane. In the second operating mode, the 4C or 4C+ interface of the RAID chip 500 is inserted into the RAID chip 500 connector on the hard drive backplane. This connector is composed of three first interfaces 110.

[0060] Furthermore, it should be noted that the E3.S interface, as a high-performance NVMe SSD interface, has a form factor similar to the conventional 2.5-inch U.2 / U.3 interface hard drives, which gives it excellent physical size compatibility. Therefore, the hard drive backplane can be flexibly designed according to actual needs to support different numbers and configurations of E3.S hard drives.

[0061] Since the E3.S interface supports multiple bandwidth options such as x2, x4, x8, and x16, the appropriate bandwidth configuration can be selected according to performance requirements.

[0062] The design of hard drive backplanes typically includes fixed locations for installing E3.S form factor RAID chips 500 (such as 16i / 32i) of different heights. These locations are carefully planned to ensure that the hard drive 300 cables in different locations are as short as possible, thereby reducing signal attenuation and interference and ensuring signal integrity.

[0063] When an E3.S RAID chip 500 is connected, these reserved positions can be directly used to connect E3.S SSDs 300. These SSDs 300 can be directly connected to the motherboard 400 slots for direct data transfer. This design ensures system flexibility while improving the utilization of hardware resources.

[0064] When connected to the E3.S RAID 500 chip, the RAID 500 can automatically identify the uplink and downlink and isolate them from the host. This automatic identification function simplifies the system configuration process and reduces the risk of human error. At the same time, the isolation function ensures that data transmission between the RAID 500 chip and the host is not interfered with by other components, improving system stability and reliability.

[0065] The RAID 500 chip also features LED functionality, achieved by sending relevant signals to the backplane CPLD (Complex Programmable Logic Device). The CPLD receives and interprets these signals, controlling the LEDs on the hard drive backplane to display different status information, such as OK, Locate, Fault, and Rebuild. This status information provides system administrators with intuitive monitoring tools, helping to promptly identify and resolve problems.

[0066] In summary, the E3.S interface hard drive 300, combined with the RAID chip 500, offers high flexibility and scalability. The hard drive backplane design fully considers factors such as hardware resource utilization, signal integrity, and system stability. Furthermore, the RAID chip 500's automatic identification and isolation functions, as well as its LED illumination function, further enhance the system's usability and maintainability.

[0067] In some embodiments, a motherboard connector is provided on the hard drive backplane, and a motherboard 400 is connected to the motherboard connector. The RAID chip 500 is electrically connected to the motherboard connector via a first connector 100 and a second connector 200. The advantage of this design is that it reduces additional cables and connection points, thereby reducing the risk of signal attenuation and interference, and improving system stability and reliability. Simultaneously, the RAID chip 500's electrical connection to the motherboard connector via the first connector 100 and the second connector 200 enables fast data transmission and efficient management.

[0068] Alternatively, a cable can be used to connect the motherboard 400 and the hard drive backplane, allowing the RAID chip 500 to be electrically connected to the cable via both the first connector 100 and the second connector 200. This design reduces overall cost. Through the cable connection, the RAID chip 500 can also be electrically connected to the cable via both the first connector 100 and the second connector 200, enabling data transmission and management.

[0069] In practical applications, the choice of connection method depends on the specific application scenario and requirements. If the system has high requirements for stability and reliability, and the frequency of hardware replacement and upgrades is low, then a direct connection (motherboard connector method) may be more suitable. If the system requires high flexibility, or the frequency of hardware replacement and upgrades is high, then a cable connection may be more suitable.

[0070] In some embodiments, such as Figure 7 and Figure 8 As shown, multiple first connectors 100 and multiple second connectors 200 are arranged in an array, with the first connectors 100 located at one end or the center of the array.

[0071] In this embodiment, the array arrangement ensures a relatively uniform physical distance between the multiple first connectors 100 and the multiple second connectors 200, which helps reduce signal attenuation and interference during transmission, improving data transmission efficiency and stability. Arranging the multiple first connectors 100 and multiple second connectors 200 in an array format makes modular design easier to achieve. Different components (such as the RAID chip 500, hard drive 300, etc.) can be inserted into different positions in the array as needed, thereby improving the system's scalability and flexibility. Furthermore, the array arrangement makes the physical locations of the connectors and components clearer and more orderly, facilitating daily management and maintenance by system administrators. For example, when a component needs to be replaced or upgraded, the corresponding location can be quickly located for operation.

[0072] With multiple first connectors 100 located at one end of the array, the overall layout is simplified, and the location of the RAID chip 500 can be easily found. Placing the first connectors 100 in the center of the array may help achieve more balanced data transmission. In this layout, the number of hard drives 300 connected to each second connector 200 on both sides of the multiple first connectors 100 is roughly equal, which helps reduce signal transmission differences and improve the overall system performance.

[0073] Specifically, such as Figure 7 As shown, this hard drive backplane design can accommodate eight 2.5-inch hard drives 300. In this scenario, it features six second interfaces 210 and two first interfaces 110. The first four second interfaces 210 are used to house PCIe 5.0 x4 NVMe SSDs with E3.S interfaces, while the latter two interfaces 110 are reserved for E3.S form factor RAID chips 500. This design ensures sufficient storage capacity while reserving expansion space to support more advanced data management functions. After inserting the E3.S form factor RAID chips 500 into the latter two reserved positions, the RAID chips 500 will manage the first four PCIe 5.0 x4 NVMe SSDs with E3.S interfaces. The RAID chips 500 can recognize the hard drives 300 on the backplane and, through their internal logic, construct various RAID arrays (such as RAID 0, 1, 5, 10, etc.) to provide data redundancy, improve performance, or balance both. The isolation between the RAID chip 500 and the host ensures that the performance of the NVMe SSDs does not fluctuate due to changes in CPU utilization. This is because the RAID chip 500 independently handles data read and write requests and communicates directly with the hard drives 300, reducing CPU intervention. Furthermore, this isolation also achieves fault isolation, ensuring that the failure of a single NVMe SSD will not affect the entire system. Since the RAID chip 500 manages the NVMe SSDs, there is no need to configure interrupts separately for each NVMe SSD. The RAID chip 500 has integrated interrupt management and error handling mechanisms, which can more effectively handle concurrent requests and abnormal situations from multiple hard drives 300.

[0074] After normal boot, the RAID chip 500 automatically identifies the E3.S hard drive 300 on the same hard drive backplane and isolates it from the host. The host manages these four isolated E3.S NVMe drives through the RAID chip 500. This ensures that their performance will not fluctuate due to changes in CPU utilization and also achieves fault isolation without the need to configure interrupts for each NVMe drive.

[0075] In this embodiment, the E3.S interface hard drives 300 are directly isolated from the host (motherboard 400) system via the RAID chip 500, ensuring that the performance of these hard drives 300 is not affected by changes in CPU utilization. The RAID chip 500 typically has its own processing unit and caching mechanism, enabling it to directly manage the read and write operations of the hard drives 300, reducing CPU intervention and thus improving overall system performance.

[0076] When hard drives 300 are managed by RAID chip 500, the failure of any single hard drive 300 will not immediately affect the operation of the entire system, because RAID chip 500 can perform data redundancy and recovery according to the configured RAID level. This architecture also simplifies the replacement and repair process of failed hard drives 300. Furthermore, since RAID chip 500 directly handles the I / O requests of hard drives 300, it reduces the number of interrupts that the CPU needs to process. In high-performance computing or large-scale data storage environments, reducing the CPU's interrupt handling burden can significantly improve system responsiveness and stability.

[0077] Specifically, such as Figure 8 As shown, this hard drive backplane design can accommodate 25 (5*5) 2.5-inch hard drives 300, arranged in a square layout of 5 rows and 5 columns. In this scenario, there are twenty-two second interfaces 210 and three first interfaces 110, with the first interfaces 110 positioned between the twenty-two second interfaces 210. The first eleven second interfaces 210 are used to house PCIe 5.0 x4 NVMe SSDs with E3.S interfaces, while the latter eleven are used to house PCIe 5.0 x4 NVMe SSDs with E3.S interfaces. The three middle first interfaces 110 are reserved for an E3.S RAID chip 500. The first interface 110 closest to the front is used for electrical connection to the eleven SSDs connected to the front, and the first interface 110 closest to the rear is used for electrical connection to the eleven SSDs connected to the rear. The first interface 110 is also electrically connected to the motherboard 400. The isolation between the RAID chip 500 and the host in this embodiment ensures the performance stability of the SSDs and achieves fault isolation. The host manages the SSDs through the RAID 500 chip, eliminating the need to directly handle interrupts and requests for each SSD, thus reducing the CPU load and improving system efficiency.

[0078] In some embodiments, such as Figure 7As shown, the hard disk backplane has multiple backplanes 600, each backplane 600 having multiple first connectors 100 and multiple second connectors 200. Each backplane 600 can switch between a first operating mode and a second operating mode. In the first operating mode, each first connector 100 and each second connector 200 within any backplane 600 is used to electrically connect to different hard disks 300. In the second operating mode, at least two first connectors 100 within any backplane 600 cooperate to jointly connect to R. The AID chip 500 and other first connectors 100 or second connectors 200 within the backplane 600 are used to electrically connect different hard drives 300. The RAID chip 500 within the backplane 600 communicates with the corresponding hard drive 300 through the corresponding first connector 100. The RAID chip 500 within the backplane 600 also communicates with the motherboard 400 through the corresponding first connector 100. The RAID chip 500 is configured to manage each hard drive 300 within the backplane 600 based on the hard drive information read from each hard drive 300 within the backplane 600.

[0079] Specifically, in this embodiment, there are three backplates 600, and each backplate 600 can switch between a first working mode and a second working mode.

[0080] When the backplane 600 is in the first operating mode, each first connector 100 and each second connector 200 within the backplane 600 are used to electrically connect to different hard drives 300. All hard drives 300 are directly connected to the motherboard 400 or connected to the motherboard 400 through the backplane, without the intervention of a RAID chip 500, making it suitable for simple storage scenarios that do not require data redundancy or high performance.

[0081] When the backplane 600 is in its second operating mode, multiple first connectors 1 within the backplane 600 work together to connect to the RAID chip 500. Meanwhile, the other first connectors 100 and second connectors 200 within the backplane 600 continue to electrically connect to different hard drives 300. At this time, the RAID chip 500 communicates with each hard drive 300 through its corresponding first connector 100 and with the motherboard 400 through the first connectors 100 within the backplane 600, enabling advanced management functions for the hard drives 300. For example, the RAID chip 500 can read hard drive information from each hard drive 300 within the backplane 600, including capacity, type, and status, which is fundamental for configuring and managing the hard drives 300. Based on the read hard drive information, the RAID chip 500 can configure and manage each hard drive 300 within the backplane 600. For example, it can create RAID arrays (such as RAID 0, 1, 5, 10, etc.) to provide data redundancy, improve performance, or balance both. Furthermore, the RAID chip 500 not only handles data communication between the hard drives 300, but also acts as a communication bridge between the hard drives 300 and the motherboard 400. It can transmit data from the hard drives 300 to the motherboard 400, and it can also transmit control commands from the motherboard 400 to the hard drives 300.

[0082] In some embodiments, such as Figure 1 and Figure 2 As shown, the hard drive backplane also includes a switching circuit. The switching circuit is configured to control the hard drive backplane to switch between a first operating mode and a second operating mode.

[0083] In this embodiment, the switching circuit can quickly and accurately control the switching of the hard disk backplane's working mode according to system requirements or administrator instructions.

[0084] In this embodiment, the switching circuit is typically integrated with monitoring and management software. It can acquire real-time status information of the hard drive backplane and connected hard drives 300, such as temperature, rotational speed, read / write speed, and error logs, and intelligently determine whether to switch operating modes based on this information. When a RAID chip 500 is detected, the switching circuit can quickly switch the affected backplane to the second operating mode, using the RAID chip 500 for data redundancy and recovery operations, thereby preventing data loss and ensuring continuous system operation. When no RAID chip 500 is detected, the switching circuit can switch more backplanes to the first operating mode, allowing the hard drives 300 to communicate directly with the motherboard 400, reducing data latency and loss in intermediate stages.

[0085] In some embodiments, the switching circuit incorporates intelligent sensors and a processor, enabling real-time monitoring of the status of the hard drive backplane and the connected hard drive 300, including key indicators such as temperature, humidity, vibration, read / write speed, and error rate.

[0086] Based on real-time monitoring data, the switching circuit uses built-in intelligent algorithms to quickly analyze and make decisions to determine the optimal operating mode.

[0087] When the system detects an abnormality in the hard drive backplane or hard drive 300 (such as performance degradation, approaching the end of its lifespan, or error codes), the switching circuit will automatically trigger a mode switch, switching the affected backplane from the first working mode to the second working mode, and using the RAID chip 500 for data protection and recovery.

[0088] At the same time, the switching circuit can dynamically adjust the working mode according to the system load and performance requirements to ensure that the storage system always operates in the optimal state.

[0089] In addition to automatic switching, the switching circuit also supports manual control and remote management. Administrators can directly control the switching circuit through the management interface or API, enabling more flexible system configuration and management. The remote management function allows administrators to monitor and manage the hard drive backplane system from any location, at any time, via a network connection, improving system maintainability and availability.

[0090] In the event of a hard drive failure (300), the switching circuit can quickly isolate the failed hard drive (300) and reassemble the remaining hard drives (300) into a new RAID array via the RAID chip (500), ensuring data security and accessibility.

[0091] The switching circuitry intelligently allocates hard drive bandwidth and cache resources based on system performance requirements and resource utilization, optimizing data transfer efficiency and response speed. By dynamically adjusting the operating modes and resource configurations of different backplanes, the switching circuitry ensures that critical tasks receive sufficient resource support while avoiding resource waste and bottlenecks.

[0092] In addition, the switching circuit also supports switching the entire hard drive backplane to safe mode under specific circumstances (such as system maintenance, upgrades, etc.) to restrict access to and modification of the hard drive 300 and ensure data security.

[0093] In one example, the hard drive backplane also includes a toggle switch and an indicator light. The toggle switch, electrically connected to the switching circuit, is integrated on the hard drive backplane and configured to receive user input or system commands to trigger the switching of the operating mode; the indicator light, electrically connected to the switching circuit, is used to display the operating mode status of the hard drive backplane.

[0094] Typically, the switch is integrated into the upper part of the hard drive back panel or an easily accessible location for convenient operation by users or system administrators. This plug-and-play design allows users to quickly switch operating modes without opening the chassis or performing complex settings.

[0095] The toggle switch is electrically connected to the switching circuit and triggers the switching of operating modes by receiving user input or system commands. Users can send switching signals through physical operations (such as toggling a switch or pressing a button), while commands may come from BIOS settings on the motherboard 400, drivers in the operating system, or remote management interfaces. The introduction of the toggle switch greatly improves the flexibility and user-friendliness of the hard drive backplane. Users can switch operating modes at any time according to actual needs without worrying about complex configuration processes or potential data loss risks.

[0096] Indicator lights are also integrated into the hard drive back panel, usually located near the toggle switch or in an easily visible position. The indicator lights should be designed to be sufficiently conspicuous to clearly display the current operating mode status in different working environments.

[0097] The indicator light is electrically connected to the switching circuit and is used to display the current operating mode status of the hard drive backplane. Depending on the status of the switching circuit, the indicator light will light up, turn off, or change color to intuitively reflect whether the hard drive backplane is in the first operating mode (direct communication mode) or the second operating mode (RAID chip 500 management mode).

[0098] The indicator lights provide users with immediate status feedback. Users don't need to navigate to system settings or view complex log information; they can quickly understand the hard drive backplane's operational status simply by observing the indicator light's color or flashing pattern. This not only improves system usability but also reduces the risk of data loss or system failure due to accidental operation.

[0099] In another embodiment of the present invention, a server is provided, comprising a hard disk backplane, a RAID chip 500, and a motherboard 400. The RAID chip 500 is communicatively connected to the motherboard 400 via a first connector 100. The structure of the hard disk backplane can be referred to the above embodiments and will not be repeated here.

[0100] Specifically, such as Figure 1 As shown, when the hard drive backplane switches to the first operating mode, in this mode, all the first connectors 100 and all the second connectors 200 on the hard drive backplane are independently used to connect different hard drives 300. Each hard drive 300 communicates directly with the motherboard 400 through its corresponding connector (first connector 100 or second connector 200) to achieve data reading, writing, and status feedback. This mode is suitable for application scenarios that do not require complex data redundancy or advanced fault isolation functions, such as single-hard drive 300 storage systems or simple data backup systems.

[0101] When the hard drive backplane switches to the second operating mode, such as Figure 2As shown, in this server mode, the operation of the hard drive backplane changes significantly. The first connector 100 no longer connects directly to the hard drive 300, but instead connects to the RAID chip 500. The RAID chip 500 is a device specifically designed for data redundancy and fault recovery, capable of managing multiple hard drives 300 to improve data security. At least two first connectors 100 are connected to the RAID chip 500, enabling the RAID chip 500 to communicate directly with some of the hard drives 300, managing their data read / write operations and fault detection. The remaining first connectors 100 and second connectors 200 continue to connect to other hard drives 300, but the data processing and management of these hard drives 300 are all handled by the RAID chip 500. The RAID chip 500 communicates with the motherboard 400 through the first connectors 100, reporting the status and data of all connected hard drives 300 to the motherboard 400, and also receiving instructions from the motherboard 400. The use of the RAID chip 500 isolates faulty hard drives 300, preventing them from affecting the operation of the entire storage system, thereby improving system stability and reliability. The RAID 500 chip enables redundant data storage, ensuring data integrity and accessibility even if some hard drives (300) fail. The RAID 500 chip optimizes data read / write operations, improving the overall performance of the storage system, especially when handling a large number of concurrent read / write requests. This mode is particularly suitable for applications requiring high data security, fault isolation, and redundant storage, such as servers and data centers.

[0102] The server provided in this embodiment of the invention, due to the aforementioned hard drive backplane, can flexibly switch between a first operating mode (direct communication mode) and a second operating mode (RAID chip 500 management mode). In the first operating mode, the first connector 100 and multiple second connectors 200 on the hard drive backplane are directly connected to the hard drive 300, enabling direct communication between the hard drive 300 and the motherboard 400. In the second operating mode, the multiple first connectors 100 cooperate to connect to the RAID chip 500. The RAID chip 500 then communicates with the motherboard 400 through the connected hard drive 300 and the first connectors 100, achieving efficient management and configuration of the hard drive 300. This isolates the hard drive 300 from the motherboard 400, providing fault isolation, hard drive 300 redundancy, and performance stability. The design of the hard drive backplane allows users to flexibly insert the RAID chip 500 or the hard drive 300 according to actual needs without having to consider complex issues such as cable connections. This plug-and-play design greatly improves the scalability and flexibility of the system, meeting the storage needs of different application scenarios.

[0103] Meanwhile, by integrating the RAID 500 connectivity into the hard drive backplane, the slot space and location occupied by the traditional RAID 500 are significantly saved. Integrating the RAID 500 connectivity into the hard drive backplane also reduces the need for an external RAID 500, thereby improving its heat dissipation. For the increasingly demanding performance and power consumption of the RAID 500, this design helps optimize heat dissipation, ensuring stable temperature operation and extending its lifespan. Eliminating the multiple cables between the traditional RAID 500 and the hard drive backplane avoids signal attenuation and link problems caused by excessive cable length or improper cable layout. This improvement not only ensures signal transmission quality but also reduces costs associated with using numerous cables. Furthermore, the cleaner cable layout helps optimize internal airflow, improving overall heat dissipation.

[0104] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0105] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A hard drive backplane, characterized in that, The hard disk backplane is provided with a plurality of first connectors and a plurality of second connectors. The hard disk backplane includes a switching circuit configured to control the hard disk backplane to switch between a first operating mode and a second operating mode. The first connectors are provided with a first interface, and the second connectors are provided with a second interface. The first interface is provided with a first pin and a second pin, and the second interface is provided with a second pin. In the first working mode, the second pin corresponding to the first interface of each first connector and the second pin corresponding to the second interface of each second connector are respectively used to electrically connect different hard drives, and each hard drive is connected to the motherboard through the first connector or the second connector. In the second operating mode, the first pins and second pins corresponding to the first interfaces of at least two of the first connectors are connected to the RAID chip. The second pins corresponding to the first interfaces of the other first connectors and the second pins corresponding to the second interfaces of all the second connectors are used to electrically connect to different hard drives. The RAID chip communicates with the corresponding hard drive through the corresponding first connector and communicates with the motherboard through the first connector. The RAID chip is configured to manage each hard drive according to the hard drive information read from each hard drive.

2. The hard disk backplane according to claim 1, characterized in that, In the second operating mode, the RAID chip is used to obtain the hard drive information of the multiple hard drives connected to the hard drive backplane, receive the RAID function enable command, select the target hard drive from the multiple hard drives according to the preset RAID level and the hard drive information of the multiple hard drives, and establish an independent hard drive redundancy array for the target hard drive.

3. The hard disk backplane according to claim 1, characterized in that, The hard drive backplane is provided with a motherboard connector, and the motherboard is connected to the motherboard connector. The RAID chip is electrically connected to the motherboard connector through both the first connector and the second connector.

4. The hard disk backplane according to any one of claims 1-3, characterized in that, The hard drive backplane has multiple backplanes, each backplane has multiple first connectors and multiple second connectors, and each backplane can switch between the first working mode and the second working mode; When any of the backplanes is in the first operating mode, the first connector and each of the second connectors in each of the backplanes are respectively used to electrically connect different hard drives; When any of the backplanes is in the second operating mode, at least two of the first connectors in the backplane are connected to the RAID chip. The other first connectors or second connectors in the backplane are used to electrically connect to different hard drives. The RAID chip in the backplane is connected to the corresponding hard drive through the corresponding first connector. The RAID chip in the backplane is connected to the motherboard through the corresponding first connector. The RAID chip is configured to manage each hard drive in the backplane according to the hard drive information read from each hard drive in the backplane.

5. The hard disk backplane according to any one of claims 1-3, characterized in that, Multiple first connectors and multiple second connectors are arranged in an array.

6. The hard disk backplane according to claim 1, characterized in that, The hard drive backplane also includes: A switching switch, electrically connected to the switching circuit, is integrated on the hard disk backplane and configured to receive user input or system commands to trigger the switching of working modes. An indicator light, electrically connected to the switching circuit, is used to display the working mode status of the hard drive backplane.

7. A server, characterized in that, include: Hard drive backplane as described in any one of claims 1-6; A RAID chip and a motherboard, wherein the RAID chip is communicatively connected to the motherboard via the first connector.

Images