High-density storage server and storage control method thereof

Abstract

The invention discloses a high-density storage server and a storage control method thereof, the high-density storage server comprises four storage control nodes and eight hard disk storage nodes, and each of the four storage control nodes is integrated with a multi-core ARM SoC main control processor; each storage control node comprises a storage control node plate, a first node structure tray and an SSD acceleration cache; each hard disk storage node comprises an SATA signal driving board and a second node structure tray; each storage control node and every two hard disk storage nodes form a storage unit, and six storage hard disks of each storage unit are controlled by a storage control node plate; the whole machine is only integrated with four ARM SoC processors with low power consumption and low cost, the hardware cost and power consumption are obviously reduced, the modular storage unit design enhances the fault isolation capability and reliability of the system, the availability is improved, and the high performance and reliability are ensured while the cost of the storage server is reduced.

Description

Technical Field

[0001] The present invention relates to the field of memory technology, and in particular to a high-density storage server and a storage control method thereof. Background Art

[0002] With the rapid development of technologies such as the Internet, 5G, and artificial intelligence, the amount of global data has exploded, and data has become an important asset for enterprises and society. In application scenarios such as big data, cloud computing, and the Internet of Things, the demand for large-capacity, high-performance storage servers is increasing. Storage servers play a key role in data centers. They not only need to provide efficient and reliable storage and data access services, but also need to ensure the security and integrity of data to support the continuous operation of key business of enterprises.

[0003] Most of the existing traditional storage servers have a PC-like architecture: a large motherboard generally has an integrated dual-core CPU (single 16-core / 32-core / 48-core, etc.), and the hard disk slots are expanded through the hard disk controller chip (plug-in card) (the common 2U server supports 12 3.5-inch hard disks or 24 2.5-inch hard disks). Although this traditional storage server design can provide higher storage performance (depending on the specific hardware configuration), it is costly, power-hungry, and has low hard disk storage density per unit chassis space, resulting in high hardware costs and energy consumption per unit storage capacity of the storage server, making it difficult to effectively meet the requirements of low-cost, low-power consumption, and high storage density in mid- and low-end application scenarios, limiting its application in the mid- and low-end market. In addition, the single-point failure rate of this type of storage server is relatively high. As long as the motherboard or key components mounted on the motherboard (such as memory sticks) have a single-point failure, it will affect the abnormal access of all hard disk data in the chassis. It is not easy to maintain. Once the motherboard is damaged, replacing the motherboard will cause a long period of business interruption.

[0004] In view of this, for mid- and low-end data storage application scenarios, existing storage servers have deficiencies in hardware cost, power consumption, hard disk storage density and system reliability, and it is difficult to meet the market demand for low-cost, low-power consumption, high hard disk storage density and high-reliability storage solutions. Summary of the invention

[0005] The object of the present invention is to provide a high-density storage server and a storage control method thereof to solve the above technical problems.

[0006] To achieve this object, the present invention adopts the following technical solutions: A high-density storage server, comprising 4 storage control nodes and 8 hard disk storage nodes, wherein each of the four storage control nodes is integrated with a multi-core ARM SoC master control processor; Each of the storage control nodes includes a storage control node board and an SSD acceleration cache, the storage control node board integrates a node structure tray, and the node structure tray is provided with two tray points for accommodating a single storage hard disk; Each of the hard disk storage nodes includes a SATA signal driver board and another node structure tray; Each of the storage control nodes and every two of the hard disk storage nodes are arranged in a column and constitute a storage unit, and the six storage hard disks of each storage unit are controlled by a multi-core ARM SoC main control processor of the storage control node board.

[0007] Optionally, each of the storage control node boards further includes onboard DDR memory and eMMC storage, and the eMMC storage serves as a system disk; Among them, the SSD acceleration cache is an M.2 NVMe SSD acceleration cache, and the M.2 NVMe SSD acceleration cache is used as a data cache.

[0008] Optionally, each of the storage control node boards extends a 2.5G high-speed Ethernet network interface through a PCIe or USB 3.0 interface, and is connected to a 2.5G Ethernet switching module through a backplane for high-speed network data communication.

[0009] Optionally, each of the storage control node boards includes a PCIe 3.0 to SATA controller chip, which expands 4 SATA signals and connects them to the corresponding SATA signal driver boards of the two hard disk storage nodes through a backplane to control the 4 storage hard disks in the hard disk storage node.

[0010] Optionally, each of the SATA signal driver boards includes a SATA signal enhancement chip and a hard disk power supply control circuit, which are used to recover and rebuild the attenuated SATA signal transmitted through the backplane and drive two locally installed storage hard disks.

[0011] The present invention also provides a storage control method for a high-density storage server, which is applied to realize the storage control of the high-density storage server as described above, and the storage control method comprises: Start a multi-core ARM SoC master processor on each storage control node, load the operating system and storage management software into the DDR memory, and initialize the storage control node board; Through the storage control node board, communication is established with two local storage hard disks and four storage hard disks connected through the backplane and the SATA signal driver board, and the six storage hard disks are logically grouped into one storage unit; In each storage unit, dynamically configuring a data storage strategy based on real-time data access patterns and workload characteristics, the data storage strategy including data distribution and fault tolerance level; By using the SSD acceleration cache, the access frequency of the data in the storage unit is monitored, the cache strategy is dynamically adjusted, and the frequently accessed data is cached in the SSD acceleration cache; The working status of the storage hard disk and the storage control node is continuously monitored, and when a hard disk failure or a node abnormality is detected, fault isolation and data recovery operations are immediately performed in the corresponding storage unit.

[0012] Optionally, in each storage unit, the six storage hard disks are logically grouped into one storage unit, and a data storage strategy is dynamically configured based on real-time data access patterns and workload characteristics, specifically including: In each storage unit, hard disk parameters of each storage hard disk are obtained, wherein the hard disk parameters include capacity, rotation speed, health status, and read / write performance parameters; Logically grouping the six storage hard disks into one storage unit, establishing a unified logical volume management, and dividing the logical storage space; Through the monitoring module on the storage control node board, data access requests in the storage unit are collected in real time to obtain workload characteristic information, which includes data access frequency, read-write ratio, I / O request size and sequentiality.

[0013] Optionally, the monitoring module on the storage control node board collects data access requests in the storage unit in real time to obtain workload feature information, and then further includes: Based on the workload characteristics collected in real time, using a predictive analysis algorithm to analyze and model data access patterns, predict future access trends, and generate a data access pattern model; According to the data access mode model, select a corresponding distributed storage strategy and configure a data storage strategy; Applying the data storage strategy to distribute and configure the data in the storage unit, the configuration process includes adjusting the data striping mode, setting the location of the verification data, and configuring the update strategy of the redundant data; The hard disk parameters and data access mode of the storage unit are continuously monitored, and when it is detected that the workload characteristic changes exceed a preset threshold, the dynamic adjustment process of the data storage strategy is repeatedly executed.

[0014] Optionally, monitoring the access frequency of data in the storage unit through the SSD acceleration cache, dynamically adjusting the cache strategy, and caching the high-frequency accessed data into the SSD acceleration cache specifically includes: In each storage unit, a monitoring module of the SSD acceleration cache is started to initialize cache parameters, wherein the cache parameters include cache capacity, cache strategy and cache threshold; Monitor the access frequency and access mode of each data block in the storage unit in real time, and collect data access statistics, the statistics including the number of read and write times, the most recent access time and the data heat index of the data block; Based on the data access statistical information, the heat value of each data block is calculated, and the hotness of the data block is evaluated according to a preset algorithm, wherein the preset algorithm includes a least recently used algorithm, a frequency counting algorithm, or a weighted moving average algorithm.

[0015] Optionally, based on the data access statistical information, a heat value of each data block is calculated, and the hotness of the data block is evaluated according to a preset algorithm, and then the following steps are further included: According to the heat value of the data block, the cache strategy is dynamically adjusted, and for the frequently accessed data blocks whose heat value exceeds the cache threshold, they are cached in the SSD acceleration cache, and for the infrequently accessed data blocks whose heat value is lower than the preset threshold, they are removed from the SSD acceleration cache; In the process of caching data, the cached data blocks are prioritized based on the capacity limitation and life characteristics of the SSD acceleration cache; When the data access mode in the storage unit changes, the dynamic update process of the cache strategy is repeatedly executed.

[0016] Compared with the prior art, the present invention has the following beneficial effects: the high-density storage server adopts a distributed and centralized architecture, including 4 storage control nodes and 8 hard disk storage nodes, each of the four storage control nodes is integrated with a multi-core ARM SoC master processor, each storage control node controls 2 local storage hard disks through its storage control node board, and is connected to two hard disk storage nodes through a SATA interface, each storage control node and its associated two hard disk storage nodes together form a storage unit, which controls data access to 6 storage hard disks in total, and the storage control node uses the ARM SoC master processor and SSD acceleration cache to process and manage data on the hard disks on the local and associated hard disk storage nodes, thereby achieving efficient data storage and reading; the whole machine integrates only 4 low-power, low-cost ARM SoC processors, which significantly reduces hardware cost and power consumption, and the modular storage unit design enhances the system's fault isolation capability and reliability, and improves availability; the storage server reduces costs while ensuring high performance and high reliability. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0018] The structures, proportions, sizes, etc. illustrated in the drawings of this specification are only used to match the contents disclosed in the specification so as to facilitate understanding and reading by persons familiar with this technology. They are not used to limit the conditions under which the present invention can be implemented, and therefore have no substantive technical significance. Any structural modification, change in proportion or adjustment of size, without affecting the effects and purposes that can be achieved by the present invention, should still fall within the scope of the technical contents disclosed by the present invention.

[0019] Figure 1 This is a schematic diagram of the system structure of the high-density storage server of the first embodiment of the present invention; Figure 2 This is a schematic diagram of the layout of the storage control nodes and hard disk storage nodes of the high-density storage server of the first embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of a storage control node of a high-density storage server of the first embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of a hard disk storage node of a high-density storage server of the first embodiment of the present invention.

[0020] Figure 5 This is a logic block diagram of the hardware system of the high-density storage server of the first embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] In the description of the present invention, it should be understood that the terms "upper", "lower", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention. It should be noted that when a component is considered to be "connected" to another component, it may be directly connected to the other component or there may be a centrally arranged component at the same time.

[0023] The technical solution of the present invention is further described below with reference to the accompanying drawings and through specific implementation methods.

[0024] Embodiment 1: Combination Figures 1 to 5 As shown, an embodiment of the present invention provides a high-density storage server, including 4 storage control nodes 20 and 8 hard disk storage nodes 30, and the four storage control nodes 20 are respectively integrated with a multi-core ARM SoC master control processor 10; each storage control node 20 includes a storage control node board 21 and an SSD acceleration cache 23, and the storage control node board 21 is provided with a node structure tray 22, and the first node structure tray 22 is provided with two tray points for accommodating a single storage hard disk, wherein the multi-core ARM SoC master control processor 10 is integrated in the storage control node board 21.

[0025] Each hard disk storage node 30 includes a SATA signal driver board 31 and another node structure tray 22, wherein each storage control node 20 and every two hard disk storage nodes 30 form a storage unit. At this time, the storage unit can accommodate 6 storage hard disks, and the 6 storage hard disks of each storage unit are controlled by a multi-core ARM SoC master processor.

[0026] The working principle of the present invention is as follows: the high-density storage server adopts a distributed and centralized architecture, including 4 storage control nodes 20 and 8 hard disk storage nodes 30. The four storage control nodes are respectively integrated with a multi-core ARM SoC master processor. Each storage control node 20 controls 2 local storage hard disks through its storage control node board 21, and is connected to two hard disk storage nodes 30 through a SATA interface. Each storage control node 20 and its associated two hard disk storage nodes 30 together form a storage unit, which controls data access to 6 storage hard disks in total. The storage control node 20 uses the ARM SoC master processor and SSD acceleration cache 23 to process and manage data on the local and associated hard disk storage nodes 30, thereby achieving efficient data storage and reading. The whole machine only integrates 4 low-power, low-cost ARM SoC processors, which significantly reduces hardware cost and power consumption. The modular storage unit design enhances the system's fault isolation capability and reliability, and improves availability. The storage server reduces costs while ensuring high performance and high reliability.

[0027] It should be noted that please combine Figure 1 As shown, the storage server of the present invention divides the chassis hardware into 7 modules, each module is physically independent of each other, and is physically connected and interconnected through a backplane. The 7 modules are: chassis, storage node module (storage control node 20 and hard disk storage node 30), backplane module, BMC management module, Ethernet switching module, power module and fan module. Each hardware module adopts a hot-swappable, modular, blade-type hardware architecture design concept, and each module can be flexibly and conveniently inserted and removed from the chassis.

[0028] Combination Figure 5 , which is a logic block diagram of the hardware system of the storage server, wherein there are 12 storage nodes in total, including 4 storage control nodes 20 and 8 hard disk storage nodes 30 .

[0029] In this embodiment, each storage control node board 21 also includes onboard DDR memory and eMMC storage, and the eMMC storage serves as a system disk; wherein the SSD acceleration cache 23 is an M.2 NVMe SSD acceleration cache 23, and the M.2 NVMe SSD acceleration cache 23 serves as a data cache.

[0030] By integrating DDR memory on the storage control node board 21, high-bandwidth, low-latency running memory is provided for the multi-core ARM SoC main control processor, supporting the efficient operation of the operating system and storage management software. eMMC storage is used as a system disk to store the operating system and key system files, with the characteristics of small size and high reliability. M.2 NVMeSSD is used as a data cache, and its high-speed read and write performance and low latency characteristics are used to cache frequently accessed data.

[0031] In this embodiment, each storage control node board 21 extends a 2.5G high-speed Ethernet network interface through a PCIe or USB 3.0 interface, and is connected to a 2.5G Ethernet switching module through a backplane 40 for high-speed network data communication.

[0032] It should be noted that by using PCIe or USB 3.0 interface to expand high-bandwidth network interface, the storage node's demand for high-speed network communication is met. The network interface is connected to the Ethernet switch module through the backplane 40, which simplifies network wiring and reduces system complexity.

[0033] In this embodiment, each storage control node board 21 includes a PCIe 3.0 to SATA controller chip, which expands 4 SATA signals and connects to the SATA signal driver boards 31 of the corresponding two hard disk storage nodes 30 through the backplane 40 to control the 4 storage hard disks in the hard disk storage node 30.

[0034] It should be noted that by integrating the PCIe 3.0 to SATA controller on the storage control node board 21, the conversion between the high-speed PCIe bus and the SATA interface is realized, and an additional SATA port is expanded. Through the connection of the backplane 40, the storage control node 20 can directly control the hard disk on the hard disk storage node 30. This design not only fully utilizes the high-speed transmission capability of PCIe3.0, but also simplifies the physical connection and reduces the complexity of wiring.

[0035] In this embodiment, it is further explained that each SATA signal driver board 31 includes a SATA signal enhancement chip and a hard disk power supply control circuit, which is used to recover and rebuild the attenuated SATA signal transmitted through the backplane 40 and drive two locally installed storage hard disks.

[0036] It should be noted that, since the SATA signal will be attenuated during the long-distance transmission of the backplane 40, the signal quality will be reduced, which may affect the stability of data transmission. By integrating the SATA signal enhancement chip on the SATA signal driver board, the attenuated signal is restored and shaped to ensure the integrity and reliability of the signal.

[0037] Embodiment 2: The present invention also provides a storage control method for a high-density storage server, which is applied to implement the storage control of the high-density storage server as in the first embodiment. The storage control method includes: S1, start the multi-core ARM SoC master processor 10 on each storage control node 20, load the operating system and storage management software into the DDR memory, and initialize the storage control node board 21; by initializing the storage control node board 21, the necessary software and hardware environment is established to ensure that the storage control node 20 can operate normally.

[0038] S2, through the storage control node board 21, establishes communication with two local storage hard disks and four storage hard disks connected through the backplane 40 and the SATA signal driver board 31, and logically groups the six storage hard disks into one storage unit; This completes the connection of all six storage hard disks in each storage unit. In this way, the storage control node 20 can fully control and manage all hard disks in the storage unit to which it belongs, and provide hardware support for data storage and access.

[0039] S3 dynamically configures data storage strategies within each storage unit based on real-time data access patterns and workload characteristics. The data storage strategies include data distribution and fault tolerance levels. Through this dynamic configuration, the system can achieve the best balance between performance, reliability, and storage efficiency to meet the needs of current workloads.

[0040] S4, monitoring the access frequency of data in the storage unit through the SSD acceleration cache 23, dynamically adjusting the cache strategy, and caching the frequently accessed data into the SSD acceleration cache 23; The system monitors the access frequency of each data block in the storage unit through the SSD acceleration cache module. According to the real-time monitoring data, the cache strategy is dynamically adjusted. For data with high access frequency, it is cached in the high-speed SSD acceleration cache to speed up data reading. For data with low access frequency, it can be removed from the cache to free up cache space.

[0041] In addition, when there is a sudden large amount of data being written, the data can be written to the SSD cache first (the writing speed of the SSD cache is faster than that of the mechanical hard disk), so the writing speed can be guaranteed. After the burst transmission is completed, the CPU will move the data in the cache to the hard disk, thereby coping well with the sudden large amount of data.

[0042] S5, continuously monitor the working status of the storage hard disk and the storage control node 20, and when a hard disk failure or node abnormality is detected, immediately perform fault isolation and data recovery operations in the corresponding storage unit.

[0043] For example, the faulty hard disk can be isolated from the storage unit, data reconstruction can be started, or data can be restored from backup. Through the fast fault handling mechanism, the integrity and availability of data can be ensured, and the impact of the fault on system operation can be reduced.

[0044] In this embodiment, it is further explained that step S3 specifically includes: S31, in each storage unit, obtaining hard disk parameters of each storage hard disk, the hard disk parameters including capacity, rotation speed, health status and read / write performance parameters; The capacity and rotation speed determine the storage capacity and read / write speed of the hard disk; health status indicators (such as SMART information) are used to evaluate the reliability of the hard disk and whether there are potential failures; read / write performance parameters reflect the current working efficiency of the hard disk. This information provides basic data for subsequent data storage strategy configuration.

[0045] S32, logically grouping the six storage hard disks into one storage unit, establishing a unified logical volume management, and dividing the logical storage space; Integrate six physically independent hard disks into one logical storage unit. Through logical volume management (LVM) technology, unified management and scheduling of storage resources are achieved, allowing the creation of flexible logical volumes on physical hard disks, facilitating the expansion and adjustment of storage space, and improving the utilization of storage resources.

[0046] S33, through the monitoring module on the storage control node board 21, real-time data access requests in the storage unit are collected to obtain workload characteristic information, which includes data access frequency, read-write ratio, I / O request size and sequentiality.

[0047] The monitoring module continuously monitors the I / O activity of the storage unit and records the detailed information of data access. The frequency of data access can identify hot data; the size and order of I / O requests (sequential access or random access) affect the setting of data striping and prefetching strategies.

[0048] S34, based on the workload characteristics collected in real time, using the predictive analysis algorithm to analyze and model the data access pattern, predict future access trends, and generate a data access pattern model; By performing statistical analysis on the collected workload characteristics and using machine learning or statistical models, data access trends can be predicted. For example, time series analysis can be used to predict the changes in the access volume of certain data in a specific time period, so that storage strategies can be adjusted in advance to meet future performance requirements.

[0049] S35. Select the corresponding distributed storage strategy according to the data access pattern model. The distributed storage strategy includes selecting the number of data replicas, erasure code parameters, and data placement strategy, and configuring the data storage strategy to achieve an optimized balance among performance, capacity, and reliability.

[0050] S36. Apply the data storage strategy to distribute and configure the data in the storage unit. The configuration process includes adjusting the data striping method, setting the position of the parity data, and configuring the update strategy for redundant data. Implement the layout of the data on the hard disk according to the selected storage strategy. For example, adjusting the data striping method can change the distribution of the data across multiple hard disks, improving the parallel read / write performance; setting the position of the parity data to ensure the reliability of the data; and configuring the update strategy for redundant data to reduce the performance overhead caused by write operations.

[0051] S37. Continuously monitor the hard disk parameters and data access pattern of the storage unit. When it is detected that the change in the workload characteristics exceeds the preset threshold, repeat the dynamic adjustment process of the data storage strategy.

[0052] Through continuous monitoring, when significant changes are found in the data access pattern, hard disk performance, or health status, re-evaluate and adjust the data storage strategy to ensure that the system continues to operate in the best state.

[0053] In this embodiment, it is further illustrated that step S4 specifically includes: S41. In each storage unit, start the monitoring module of the SSD acceleration cache 23 and initialize the cache parameters. The cache parameters include cache capacity, cache policy, and cache threshold.

[0054] S42. Real-time monitor the access frequency and access pattern of each data block in the storage unit, collect data access statistical information. The statistical information includes the read / write times of the data block, the most recent access time, and the data popularity metric.

[0055] S43. Based on the data access statistical information, calculate the popularity value of each data block, and evaluate the hotness degree of the data block according to the preset algorithm. The preset algorithms include the Least Recently Used (LRU) algorithm, frequency counting algorithm, or weighted moving average algorithm.

[0056] S44. According to the popularity value of the data block, dynamically adjust the cache policy. For the frequently accessed data blocks whose popularity value exceeds the cache threshold, cache them in the SSD acceleration cache 23. For the infrequently accessed data blocks whose popularity value is lower than the preset threshold, remove them from the SSD acceleration cache 23. S45, in the process of caching data, the cached data blocks are prioritized based on the capacity limit and life characteristics of the SSD acceleration cache 23; data blocks with high read frequency and moderate size are cached first to avoid frequent writing of small data blocks causing excessive wear of the SSD; Continuously monitor the performance indicators of the SSD acceleration cache 23, including cache hit rate, cache response time, and cache space utilization. When it is detected that the performance indicators are lower than the preset standards, adjust the cache parameters, including increasing the cache capacity, modifying the cache replacement strategy, or adjusting the cache pre-fetching mechanism, to optimize the cache performance.

[0057] S46, when the data access pattern in the storage unit changes, the dynamic update process of the cache strategy is repeatedly executed to ensure that the frequently accessed data always receives a quick response, thereby improving the overall data processing efficiency of the system.

[0058] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A high-density storage server, characterized in that: It includes 4 storage control nodes and 8 hard disk storage nodes, and each of the four storage control nodes is integrated with a multi-core ARM SoC master control processor; Each of the storage control nodes includes a storage control node board and an SSD acceleration cache, wherein the storage control node board is integrated with a node structure tray, and the node structure tray is provided with two tray points for accommodating a single storage hard disk; Each of the hard disk storage nodes includes a SATA signal driver board and another node structure tray; Each of the storage control nodes and every two of the hard disk storage nodes are arranged in a column and constitute a storage unit, and the six storage hard disks of each storage unit are controlled by a multi-core ARM SoC main control processor of the storage control node board.

2. The high-density storage server according to claim 1, characterized in that: Each of the storage control node boards also includes onboard DDR memory and eMMC storage, and the eMMC storage serves as a system disk; Among them, the SSD acceleration cache is an M.2 NVMe SSD acceleration cache, and the M.2 NVMe SSD acceleration cache is used as a data cache.

3. The high-density storage server according to claim 2, characterized in that: Each of the storage control node boards extends a 2.5G high-speed Ethernet network interface through a PCIe or USB 3.0 interface, and is connected to a 2.5G Ethernet switching module through a backplane for high-speed network data communication.

4. The high-density storage server according to claim 2, characterized in that: Each of the storage control node boards includes a PCIe 3.0 to SATA controller chip, which expands 4 SATA signals and connects to the corresponding SATA signal driver boards of the two hard disk storage nodes through the backplane to control the 4 storage hard disks in the hard disk storage node.

5. The high-density storage server according to claim 4, characterized in that: Each of the SATA signal driver boards includes a SATA signal enhancement chip and a hard disk power supply control circuit, which is used to recover and rebuild the attenuated SATA signal transmitted through the backplane and drive two locally installed storage hard disks.

6. A storage control method for a high-density storage server, characterized in that: Applied to implement storage control of a high-density storage server according to any one of claims 1 to 5, the storage control method comprising: Start a multi-core ARM SoC master processor on each storage control node, load the operating system and storage management software into the DDR memory, and initialize the storage control node board; Through the storage control node board, communication is established with two local storage hard disks and four storage hard disks connected through the backplane and the SATA signal driver board, and the six storage hard disks are logically grouped into one storage unit; In each storage unit, dynamically configuring a data storage strategy based on real-time data access patterns and workload characteristics, the data storage strategy including data distribution and fault tolerance level; By using the SSD acceleration cache, the access frequency of the data in the storage unit is monitored, the cache strategy is dynamically adjusted, and the frequently accessed data is cached in the SSD acceleration cache; The working status of the storage hard disk and the storage control node is continuously monitored, and when a hard disk failure or a node abnormality is detected, fault isolation and data recovery operations are immediately performed in the corresponding storage unit.

7. The storage control method of the high-density storage server according to claim 6, characterized in that: In each storage unit, the six storage hard disks are logically grouped into one storage unit, and a data storage strategy is dynamically configured based on real-time data access patterns and workload characteristics, specifically including: In each storage unit, hard disk parameters of each storage hard disk are obtained, wherein the hard disk parameters include capacity, rotation speed, health status, and read / write performance parameters; Logically grouping the six storage hard disks into one storage unit, establishing a unified logical volume management, and dividing the logical storage space; Through the monitoring module on the storage control node board, data access requests in the storage unit are collected in real time to obtain workload characteristic information, which includes data access frequency, read-write ratio, I / O request size and sequentiality.

8. The storage control method of the high-density storage server according to claim 7, characterized in that: By using the monitoring module on the storage control node board, data access requests in the storage unit are collected in real time to obtain workload feature information, and then the following steps are further included: Based on the workload characteristics collected in real time, using a predictive analysis algorithm to analyze and model data access patterns, predict future access trends, and generate a data access pattern model; According to the data access mode model, select a corresponding distributed storage strategy and configure a data storage strategy; Applying the data storage strategy to distribute and configure the data in the storage unit, the configuration process includes adjusting the data striping method, setting the location of the verification data, and configuring the update strategy of the redundant data; The hard disk parameters and data access mode of the storage unit are continuously monitored, and when it is detected that the workload characteristic changes exceed a preset threshold, the dynamic adjustment process of the data storage strategy is repeatedly executed.

9. The storage control method of the high-density storage server according to claim 6, characterized in that: The SSD acceleration cache is used to monitor the access frequency of the data in the storage unit, dynamically adjust the cache strategy, and cache the high-frequency access data into the SSD acceleration cache, specifically including: In each storage unit, a monitoring module of the SSD acceleration cache is started to initialize cache parameters, wherein the cache parameters include cache capacity, cache strategy and cache threshold; Monitor the access frequency and access mode of each data block in the storage unit in real time, and collect data access statistics, the statistics including the number of read and write times, the most recent access time and the data heat index of the data block; Based on the data access statistical information, the heat value of each data block is calculated, and the hotness of the data block is evaluated according to a preset algorithm, wherein the preset algorithm includes a least recently used algorithm, a frequency counting algorithm, or a weighted moving average algorithm.

10. The storage control method of the high-density storage server according to claim 9, characterized in that: Based on the data access statistics, the heat value of each data block is calculated, and the hotness of the data block is evaluated according to a preset algorithm, and then the following steps are further included: Dynamically adjust the cache strategy according to the heat value of the data block, cache the frequently accessed data blocks whose heat value exceeds the cache threshold into the SSD acceleration cache, and remove the infrequently accessed data blocks whose heat value is lower than the preset threshold from the SSD acceleration cache; In the process of caching data, the cached data blocks are prioritized based on the capacity limitation and life characteristics of the SSD acceleration cache; When the data access mode in the storage unit changes, the dynamic update process of the cache strategy is repeatedly executed.

Images