A storage unit fault screening circuit, method, device, storage medium and system

By designing a memory unit fault screening circuit in memory test and using the Kulun metering unit to measure the charge, the problem of difficult to take into account in the existing technology of testing efficiency and coverage is solved, and fast and accurate memory fault location and performance evaluation are achieved. Testing servers with large capacity memory.

CN118711650BActive Publication Date: 2024-11-01BEIJING JINQUN TECH CO LTD
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202411196039.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-29
Publication Date
2024-11-01
Estimated Expiration
2044-08-29

AI Technical Summary

Technical Problem

Existing memory testing technologies are difficult to take into account test coverage and testing efficiency, and it is difficult to quickly and accurately locate memory failure locations and performance data.

Method used

A memory cell fault screening circuit is designed, including a control unit, a switching unit and a Coulomb metering unit. By separately controlling the discharge of the memory array and measuring the charge amount using the Coulomb metering unit, it realizes the evaluation and failure of the performance level of the memory chip. Accurate positioning of intervals.

Benefits of technology

It realizes high-precision and fast memory failure location and performance evaluation, improves testing efficiency and reduces costs, and is suitable for factory testing of servers with large-capacity memory.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118711650B_ABST
    Figure CN118711650B_ABST
Patent Text Reader

Abstract

The present invention is applicable to the field of memory detection technology, and provides a storage unit fault screening circuit, method, device, storage medium and system, wherein the circuit includes a control unit, a switch unit and a coulomb metering unit; the control unit is connected to the switch unit to control the working state of each switch unit; the switch unit is connected to the memory to be tested and the coulomb metering unit, and a plurality of switch units are provided, each of which corresponds to a storage array; the coulomb metering unit is used to measure the charge amount of the current introduced by the switch unit. This circuit can discharge different memory arrays separately and measure the charge retention of each array, thereby accurately knowing the performance data of each array module, accurately locating the fault area and fault level, and the overall test accuracy of the device is high, the speed is fast, and the cost of generation and use is low, which is conducive to promotion and use.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of memory detection technology, and in particular relates to a storage unit fault screening circuit, method, device, storage medium and system. Background Art

[0002] The reliability and stability of servers are crucial to the normal operation of enterprises and organizations. According to statistics, among the server downtime events, the downtime caused by memory failure accounts for the largest proportion, about 70%. Therefore, before assembling the server, it is necessary to conduct strength tests on memory components, especially memory particles.

[0003] Current memory testing methods include software testing and hardware testing. In software testing, the software is used to read and write the memory chip or memory stick as a whole, and different functions are tested by changing the read and write timing and using a variety of different types of test algorithms. Hardware testing is mainly carried out by adjusting test conditions such as operating voltage and operating temperature.

[0004] The current memory testing technologies in the industry all have the problem of conflicting test coverage and test efficiency, and existing testing devices are difficult to quickly and accurately locate the fault location of the memory and the performance data of the memory chip. Summary of the invention

[0005] The purpose of the embodiments of the present application is to provide a storage unit fault screening circuit, which aims to solve the problem that the current memory testing technology in the industry has difficulty in balancing test coverage and test efficiency, and the existing testing devices are difficult to quickly and accurately locate the fault location of the memory and the performance data of the memory chip.

[0006] The embodiment of the present application is implemented as follows: a storage unit fault screening circuit is provided, wherein a plurality of storage units form a storage array, and the circuit comprises: a control unit, a switch unit, and a coulomb metering unit; the control unit is connected to the switch unit and is used to control the working state of each switch unit; the switch unit is connected to the memory to be tested and the coulomb metering unit, and a plurality of switch units are provided, each of which is connected to a corresponding storage array; the switch unit has two working states, working state one is to connect the coulomb metering unit to the storage array, and working state two is to connect the storage array to the ground line; the coulomb metering unit is used to measure the charge amount of the current introduced by the switch unit.

[0007] Preferably, the coulomb metering unit comprises: a coulomb metering chip, a measuring resistor and a grounding terminal; the first end of the measuring resistor is the input end of the coulomb metering unit, which is used to be connected to the switching unit, and the second end of the measuring resistor is grounded; the coulomb metering chip is used to measure the amount of charge flowing through the measuring resistor, and the coulomb metering chip has a positive measuring end and a negative measuring end, the positive measuring end is connected to the first end of the measuring resistor, and the negative measuring end is connected to the second end of the measuring resistor; the coulomb metering chip also has a measurement signal output end, which is used to output the coulomb measurement value.

[0008] Preferably, the switch unit includes: the switch unit includes: a first input end, a first output end, a second output end and a control end; the first input end is used to connect to a storage array; the first output end is grounded; the second output end is connected to the input end for the coulomb metering unit; the control end is connected to a control signal output end of the control unit.

[0009] Preferably, four switch units are provided, and the second output end of each switch unit is connected to a measurement bus, and the measurement bus is connected to the input end of the coulomb metering chip.

[0010] Another object of the embodiment of the present application is to provide a storage unit fault screening method, the method is implemented based on the storage unit fault screening circuit as described above, and the method comprises the following steps:

[0011] S10, connecting the measurement signal output terminal of the coulomb metering unit and the control signal input terminal of the control unit to the SMBUS bus of the memory to be tested;

[0012] S20, turning off the memory interleaving function of the memory to be tested, parsing the address range of the storage array of the memory to be tested, and locating the data chip and storage array of the memory to be tested; each memory to be tested contains a number of storage blocks, each of the storage blocks contains a number of data chips, and each of the data chips contains a number of storage arrays;

[0013] S30, powering each storage array of the memory to be tested separately and performing a write 1 operation; after a preset discharge time, writing 0 to the array, and obtaining the remaining coulomb amount of the array after writing 0 based on the coulomb metering unit;

[0014] S40, testing all the storage arrays of the memory to be tested one by one to obtain a residual coulomb set of all the storage arrays of each memory chip, and comparing the residual coulomb set with a preset performance parameter threshold to obtain a performance level of the memory chip.

[0015] Preferably, the method further comprises:

[0016] S50, based on the performance parameter threshold, screening out faulty storage chips and faulty storage arrays that meet preset fault standards;

[0017] S60, based on a memory fault test algorithm, accurately locate the fault interval in the faulty storage array to obtain an address of the fault interval;

[0018] S70: Repair the fault interval based on a memory fault repair algorithm and the address of the fault interval.

[0019] Preferably, the method further comprises:

[0020] S80. After completing step S70, repeat steps S10 to S40 for multiple times to obtain a memory chip whose remaining coulomb amount set is less than a preset value, and repair the memory chip based on steps S60 to S70 to improve the yield rate.

[0021] Another object of an embodiment of the present application is to provide a storage unit fault screening device, the storage unit fault screening device comprising:

[0022] A measurement start module, used to connect the measurement signal output terminal of the coulomb metering unit and the control signal input terminal of the control unit to the SMBUS bus of the memory to be tested;

[0023] The measurement preset value module is used to turn off the memory interleaving function of the memory to be tested, and parse the address range of the storage array of the memory to be tested, and locate the data chip and storage array of the memory to be tested; each memory to be tested contains a number of storage blocks, each of the storage blocks contains a number of data chips, and each of the data chips contains a number of storage arrays;

[0024] The remaining coulomb acquisition module is used to supply power to each storage array of the memory to be tested separately and perform a write 1 operation; after a preset discharge time, the array is written 0, and the remaining coulomb amount of the array after writing 0 is obtained based on the coulomb metering unit;

[0025] The chip performance level acquisition module is used to test all the storage arrays of the memory to be tested one by one, obtain the remaining coulomb set of all the storage arrays of each memory chip, compare the remaining coulomb set with the preset performance parameter threshold, and obtain the performance level of the memory chip.

[0026] Another object of an embodiment of the present application is to provide a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the processor executes the steps of the storage unit fault screening method as described above.

[0027] Another object of an embodiment of the present application is to provide a storage unit fault screening system, which, when running, executes the steps of the storage unit fault screening method as described above.

[0028] A storage unit fault screening circuit provided in an embodiment of the present application can individually control the discharge of different memory arrays, and obtain the charge retention status of each storage array through a coulomb metering unit, so as to accurately obtain the performance data of each array module, accurately locate the fault area and fault level, and the overall test accuracy and speed of the device are high, and the cost of generation and use is low, which is conducive to promotion and use. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A connection diagram of a storage unit fault screening circuit module provided in an embodiment of the present application;

[0030] Figure 2 A circuit diagram of a storage unit fault screening circuit provided in an embodiment of the present application;

[0031] Figure 3 A schematic diagram of a memory chip provided in an embodiment of the present application;

[0032] Figure 4 A pin connection diagram of a memory chip provided in an embodiment of the present application;

[0033] Figure 5 A schematic diagram of a switch module integrated with four switch units provided in an embodiment of the present application;

[0034] Figure 6 A flowchart of the operation steps of a storage unit fault screening method provided in an embodiment of the present application;

[0035] Figure 7 A module block diagram of a storage unit fault screening device provided in an embodiment of the present application;

[0036] Figure 8 FIG. 4 is a block diagram of the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0038] It is understood that the terms "first", "second", etc. used in this application may be used herein to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, a first end may be referred to as a second end, and similarly, a second end may be referred to as a first end without departing from the scope of this application.

[0039] like Figure 1 As shown, it is a connection schematic diagram of a storage unit fault screening circuit module provided in an embodiment of the present application. In this embodiment, a plurality of storage units form a storage array, and the circuit includes: a control unit, a switch unit and a coulomb metering unit; the control unit is connected to the switch unit and is used to control the working state of each switch unit; the switch unit is connected to the memory to be tested and the coulomb metering unit, and a plurality of switch units are provided, each of which is connected to a corresponding storage array; the switch unit has two working states, working state one is to connect the coulomb metering unit to the storage array, and working state two is to connect the storage array to the ground line; the coulomb metering unit is used to measure the charge amount of the current introduced by the switch unit.

[0040] In the embodiment of the present application, the above test circuit can be used to quickly lock potential memory fault array units in the factory test of large-capacity memory servers. Then, after the faulty unit is determined, the algorithm is used to accurately detect the fault, so that the high fault detection rate is guaranteed while greatly improving the test efficiency of large-capacity memory.

[0041] A storage unit fault screening circuit is provided through the embodiment of the present application, which can individually control the discharge of different memory arrays, and obtain the charge retention status of each storage array through the coulomb metering unit, so as to accurately know the performance data of each array module, accurately locate the fault area and fault level, and the overall test accuracy of the device is high, the speed is fast, and the cost of generation and use is low, which is conducive to promotion and use.

[0042] In one embodiment, Figure 2 As shown, a specific circuit structure diagram of a storage unit fault screening circuit is provided. Figure 2U1 is the DRAM chip, U2 is the metering chip in the coulomb metering unit, U3 is the control chip in the control unit, and U4 / U5 / U6 / U7 are the single-pole double-throw switch chips in the four switch units. The memory to be tested is a Dimm memory bar, and the grounding part of the power supply of the four storage array areas inside all DRAM particles on it is controlled by the switch chip through the control unit to control whether it flows through the coulomb meter chip U2. The IIC communication bus SDC / SCL signal of U2 / U3 is connected to the SMBUS bus on the DIMM memory PCB.

[0043] In one embodiment, Figure 3 As shown, it is a memory chip. Figure 4 The left side of the middle is a schematic diagram of the pin definition of a certain model of DRAM memory particles. Figure 4 The black dots on the right side of the middle are the solder contacts of each pin on the back of the memory chip die, corresponding to Figure 4 The name of each pin in the yellow box on the left. The internal memory cells of the memory chip are distributed in the form of an array. Figure 3 The power supply and control data signals of each area are connected to the nearest black dot in the A, B, C, and D areas. Figure 3 The narrow signal E area inside the chip; Figure 3 The internal signal PadE area of ​​the chip will connect the signal to the four areas A, B, C, and D nearby, so Figure 4 The 20 pairs of power (VDD) and ground (VSS) pins in the chip can be divided into four power pairs a / b / c / d to supply power to the four storage array areas, except for the power supply of the control and cache circuits inside the chip. In actual applications, a memory is composed of multiple ranks. The double-rank DIMM memory has two groups of chips; the quad-rank has four groups of chips, and the rank number names are sorted from Rank0 to Rank3. If an X8-bit wide chip (8-bit data output) is used, each rank is composed of a group of 9 memory chips, of which the 8 chips from 1 to 8 are data chips, and the 9th chip is an ECC check chip, with a total of 64-bit effective data bit width and 8-bit ECC data bit width.

[0044] In a preferred embodiment of the present application, the coulomb metering unit includes: a coulomb metering chip, a measuring resistor and a grounding terminal; the first end of the measuring resistor is the input end of the coulomb metering unit, which is used to be connected to the switching unit, and the second end of the measuring resistor is grounded; the coulomb metering chip is used to measure the amount of charge flowing through the measuring resistor, and the coulomb metering chip has a positive measuring end and a negative measuring end, the positive measuring end is connected to the first end of the measuring resistor, and the negative measuring end is connected to the second end of the measuring resistor; the coulomb metering chip also has a measurement signal output end, which is used to output the coulomb measurement value.

[0045] In this embodiment, if Figure 2 As shown, the measured resistor is R1, and a 50 milliohm resistor can be selected.

[0046] In a preferred embodiment of the present application, the switch unit includes: a first input terminal, a first output terminal, a second output terminal and a control terminal; the first input terminal is used to connect to a storage array; the first output terminal is grounded; the second output terminal is connected to the input terminal for the coulomb metering unit; and the control terminal is connected to a control signal output terminal of the control unit.

[0047] In this embodiment, if Figure 2 As shown in FIG. 4 , a schematic diagram of the pin connection of a switch unit chip is given. The four switch units can be integrated in Figure 5 In a switch chip shown in FIG. Figure 2 The control chip MCU shown can use 4 control output ports to control the opening and closing conditions of the four single-pole double-throw switches.

[0048] In a preferred embodiment of the present application, four switch units are provided, and the second output end of each switch unit is connected to a measurement bus, and the measurement bus is connected to the input end of the coulomb metering chip.

[0049] In this embodiment, the four switches share a coulomb counter to reduce the cost. The four switches can be controlled separately, for example, to be turned on in turn.

[0050] In one embodiment, a storage unit fault screening method is proposed. This embodiment can be applied to servers, notebooks, desktops, industrial control computers, mobile devices (mobile phones, pads, learning machines, readers, etc.), embedded devices, etc., but is not limited to this. The method described in this application does not necessarily need to be implemented based on ECC memory, but is a general memory fault detection circuit, method and device that can be used in any occasion where memory particles are applied. A storage unit fault screening method can specifically include the following steps:

[0051] S10, connecting the measurement signal output terminal of the coulomb metering unit and the control signal input terminal of the control unit to the SMBUS bus of the memory to be tested.

[0052] S20, turning off the memory interleaving function of the memory to be tested, parsing the address range of the storage array of the memory to be tested, and locating the data chip and storage array of the memory to be tested; each memory to be tested contains a number of storage blocks, each of the storage blocks contains a number of data chips, and each of the data chips contains a number of storage arrays.

[0053] S30, powering each storage array of the memory to be tested separately and performing a write 1 operation; after a preset discharge time, writing 0 to the array, and obtaining the remaining coulomb amount of the array after writing 0 based on the coulomb metering unit.

[0054] S40, testing all the storage arrays of the memory to be tested one by one to obtain a residual coulomb set of all the storage arrays of each memory chip, and comparing the residual coulomb set with a preset performance parameter threshold to obtain a performance level of the memory chip.

[0055] In this embodiment, the essence of the storage unit cell failure is the decrease in charge retention capability. A storage block can refer to a rank of memory. A storage chip can refer to a chip of memory. The design idea of ​​the embodiment of the present application is to compare the difference in physical charge consumption of each storage array in each storage chip, lock the array where the cell of the faulty memory is located, and then perform a memory read and write test on the array, and after locking the faulty cell, use, for example, dimmPPR (PostPackage Repair) technology to replace and repair it, so as to achieve efficient rapid detection and repair of the faulty memory cell.

[0056] In one embodiment, the specific steps of the test may be: after the charge refresh function of the memory cell of the memory can be turned off (all cells storing 1 in the array are recharged), the memory array is fully written with 1, i.e., the high level is fully charged, and after the natural leakage of the array (natural leakage refers to the loss of the charge of the memory cell to the VSS ground network closest to the chip) for a period of time t (e.g., 10 to 10000 ms), the charge of the memory array is counted using a coulomb counter chip that can accurately measure the charge. The action of writing all 0s to the memory array is the process of flowing the array charge through the coulomb counter chip circuit to the VSS ground. Finally, the coulomb counter chip data is read to detect the total amount of charge remaining in the array. The smaller the total amount of charge remaining, the worse the charge retention ability of the surface memory array.

[0057] In one embodiment of the present application, the specific method implementation steps are as follows:

[0058] 1. For the Dimm memory on the packaged PCB circuit board, you can first connect the four power signals a / b / c / d of each memory chip in parallel, and then connect them to four coulomb meter chips that can accurately measure the charge, and connect the data communication IIC interface of the coulomb meter chip to the SMBUS bus of the SPD chip. Connect the VDD of the four power signals a / b / c / d of the memory chip to the power supply end (the chip power supply source of DDR4 is the DIMM gold finger, and the chip power supply source of DDR5 is the PMIC circuit), and the VSS ground end of the four power signals is controlled by the control unit to be connected to the coulomb meter chip or directly grounded; the IIC interface of the control chip of the control unit is connected to the SMBUS bus of the Dimm gold finger. The full name of the SPD chip is Serial PresenceDetect, which is an erasable electronic information chip embedded in the memory module. It records various important information of the memory module, including memory chip and module manufacturers, operating frequency, operating voltage, speed, capacity, voltage and row and column address bandwidth parameters. The motherboard can read this information through the SPD chip and configure the memory work according to this information to ensure that the memory bar can run stably and efficiently. When performing a quick screening action, the four switching circuits can be controlled separately. For example, in the first round of screening, the VSS of the power supply of group a is switched to the coulomb counter, and b, c, and d are all switched to GND. The array of storage particles is written with all 1s and then all 0s. At this time, the data of a can be obtained from the coulomb counter. Similarly, when measuring b or c / d, the switch circuit corresponding to the array is turned on, and the switch circuits of the other three are turned off.

[0059] 2. Before the test begins, enter the server BIOS Setup interface and turn off the memory interleaving option function so that the operating system address space is the same as the memory physical address distribution, and all memory addresses are arranged in order according to the chip capacity. In the address range 0~4G, the memory high base address~memory maximum capacity address, two continuous memory intervals, according to the average distribution of each storage array capacity, parse the address range of each storage array, corresponding to the four storage arrays in the eight data chips of each Rank. Among them, memory interleaving refers to distributing continuous memory data on different memory modules to achieve load balancing and potential performance improvement. However, in some tests or special application environments, it may be necessary to directly access the physical address of the memory to avoid the indirect addressing caused by interleaving in order to obtain a more direct memory access mode. After turning off the interleaving function, memory access will follow the natural order of physical addresses, that is, the logical address seen by the operating system is consistent with its corresponding physical address. Mapping the memory address space means: by turning off interleaving, you can ensure that the operating system address space is the same as the memory physical address distribution, which means that the conversion of logical addresses to physical addresses becomes more direct, without offsets caused by interleaving. This is very important for testing or debugging scenarios that require precise control of memory access. Parsing the storage array address range means: in a continuous memory interval, according to the capacity of the storage array (usually the sub-unit on the memory stick), the entire address space is evenly divided, and a continuous address range is allocated to each array. This step is crucial for understanding the memory layout, locating errors, or performing specific performance tests. Locating data chips and storage arrays means: further refining the division of the address space, corresponding the address range of each storage array to a specific rank and the data chip on the rank. Each rank may contain multiple data chips, and each data chip may contain multiple storage arrays. In this way, it is possible to accurately identify which memory address belongs to which specific physical storage unit, which is very useful for low-level hardware testing and troubleshooting. In general, these operations in this step are to create a clear memory address mapping relationship, which is convenient for direct and accurate access to the specific location of the memory, so that we can directly and accurately control the memory area we want to screen, which is convenient for subsequent operations and efficiency improvement.

[0060] 3. Select the array a of all storage chips and supply power through the coulomb metering chip. Perform a full write 1 operation on the array a of Chip0 of Rank0. After the preset discharge time t, the array a is fully written to 0. When the current coulomb data of the coulomb metering chip is read before writing 0, After writing 0, read the current coulomb data of the coulomb metering chip. , we can get the remaining coulombs of the array after completing the write 0 operation:

[0061] .

[0062] 4. Repeat step 3 to calculate the four arrays of each Chip of all ranks. Value Collection After counting multiple batches of memory and troubleshooting faulty cells using the memory test algorithm on a large number of arrays with relatively low coulomb residual values, the dangerous coulomb residual threshold level is designed based on the faulty cell detection rate in each coulomb residual value interval. For example, as shown in the following table, the dangerous coulomb residual threshold can be defined as three levels, P1, P2, and P3, based on the detection rate difference, where P3 is more stringent. By comparing with the coulomb range in the table below, the performance level of the memory chip can be obtained.

[0063]

[0064] Through this method, the fault location and the performance level of the storage chip can be accurately and quickly obtained, thereby achieving efficient and accurate positioning of the fault.

[0065] In a preferred embodiment of the present application, the method further comprises:

[0066] S50: Based on the performance parameter threshold, screen out faulty storage chips and faulty storage arrays that meet preset fault standards.

[0067] S60: Based on a memory fault test algorithm, accurately locate the fault interval in the faulty storage array to obtain an address of the fault interval.

[0068] S70: Repair the fault interval based on a memory fault repair algorithm and the address of the fault interval.

[0069] In an embodiment of the present application, some storage arrays with potential faulty storage unit cells can be screened according to the threshold corresponding to the established dangerous coulomb quantity remaining threshold level, and then using memory testing software, AMT in BIOS and other testing tools with memory fault testing algorithms, a test algorithm with higher algorithm complexity (for example, an algorithm that consumes relatively more time but has a higher algorithm detection rate) can be used to further test the address interval corresponding to the potential fault array and quickly screen out the faulty cell unit. Finally, the memory fault repair algorithm and the data of the address of the fault interval are used to perform PPR repair. Generally speaking, the relationship between the test algorithm and time complexity is that within a unit of memory capacity, the more complex the pattern written by the memory test algorithm, the longer the test time, and the more read and write operation steps of the memory test algorithm, the longer the test time.

[0070] In a preferred embodiment, the method further comprises:

[0071] S80. After completing step S70, repeat steps S10 to S40 for multiple times to obtain a memory chip whose remaining coulomb amount set is less than a preset value, and repair the memory chip based on steps S60 to S70 to improve the yield rate.

[0072] In the embodiment of the present application, optionally, in order to achieve a more complete detection effect, after the PPR is repaired, the above steps S10 to S40 can be repeated multiple times, and the array value set 1~3 arrays with smaller residual coulomb values ​​are tested by the algorithm of the test tool again. After the faulty cells are detected, PPR repair is performed again. Thereby improving the yield rate. The selection of the residual coulomb threshold level can be differentiated for different customers using the server. For example, regular users can use the P1 level, financial securities users can use the P2 level, and key field users can use the P3 level. The memory fault detection method provided by this solution can provide enterprises and server manufacturers with a reliable memory fault preprocessing solution, which can efficiently and accurately predict potential failures of memory particles in advance to ensure the stability and reliability of the server, ensure the quality of server memory components after assembly, and achieve quantifiable improvement in server reliability indicators.

[0073] In one embodiment of the present application, Figure 6 As shown in FIG. 1 , a flowchart of a method for screening storage unit faults is also provided. Figure 6 As shown in the figure, when the test starts, the i-th memory is selected, and the j-th Rank and k-th memory chip (particle) are selected one by one for testing. In the BIOS, SMBUS is used to send instructions to the control unit to select the m-th array for testing. First, write all 1s, wait for natural discharge, and then write 0s. Read the charge flowing through the coulomb metering unit before and after writing 0, that is, the difference . In this way, all memory block ranks, memory chips, and memory arrays of all memories are traversed and tested one by one.

[0074] In this embodiment, based on the detection method, a large number of memory chips can be automatically tested. The test performed by the method can start from the first memory to be tested and detect all storage arrays in all memories.

[0075] In one embodiment, Figure 7 As shown, a storage unit fault screening device is provided, which can be integrated into various devices with control or computing capabilities, and specifically may include:

[0076] A measurement start module 510, used to connect the measurement signal output terminal of the coulomb metering unit and the control signal input terminal of the control unit to the SMBUS bus of the memory to be tested;

[0077] The measurement preset value module 520 is used to turn off the memory interleaving function of the memory to be tested, and parse the address range of the storage array of the memory to be tested, and locate the data chip and storage array of the memory to be tested; each memory to be tested contains a number of storage blocks, each of the storage blocks contains a number of data chips, and each of the data chips contains a number of storage arrays;

[0078] The remaining coulomb acquisition module 530 is used to supply power to each storage array of the memory to be tested separately and perform a write 1 operation; after a preset discharge time, write 0 to the array, and obtain the remaining coulomb amount of the array after writing 0 based on the coulomb metering unit;

[0079] The chip performance level acquisition module 540 is used to test all the storage arrays of the memory to be tested one by one, obtain the remaining coulomb set of all the storage arrays of each memory chip, compare the remaining coulomb set with the preset performance parameter threshold, and obtain the performance level of the memory chip.

[0080] In the embodiment of the present application, the explanation and description of the above-mentioned storage unit fault screening device can refer to the explanation and description of the above-mentioned corresponding method. For the description of the above-mentioned storage unit fault screening method, please refer to the above, which will not be repeated here.

[0081] In the embodiment of the present application, based on the present device, a reliable memory fault preprocessing solution can be provided for enterprises and server manufacturers, which can efficiently and accurately predict potential failures of memory particles in advance to ensure the stability and reliability of the server, guarantee the quality of the server memory components after assembly, and achieve a quantifiable improvement in server reliability indicators.

[0082] In one embodiment, a storage unit fault screening system is provided, and when the system is running, the steps of the storage unit fault screening method described above are executed.

[0083] In the embodiment of the present application, the system may be a computer software or program system, which runs in hardware containing a processing device, and when the system is running, the corresponding method is executed. For the description of the storage unit fault screening method, please refer to the above text, which will not be repeated here.

[0084] In the embodiments of the present application, based on the present system, a reliable memory fault preprocessing solution can be provided for enterprises and server manufacturers, which can efficiently and accurately predict potential failures of memory particles in advance to ensure the stability and reliability of the server, guarantee the quality of the server memory components after assembly, and achieve a quantifiable improvement in server reliability indicators.

[0085] Figure 8FIG. 2 shows an internal structure diagram of a computer device in one embodiment. Figure 8 As shown, the computer device includes a processor, a memory, a network interface, an input device and a display screen connected through a system bus. Among them, the memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and may also store a computer program. When the computer program is executed by the processor, the processor can implement a storage unit fault screening method. The internal memory may also store a computer program. When the computer program is executed by the processor, the processor can execute a storage unit fault screening method. The display screen of the computer device may be a liquid crystal display or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a button, trackball or touchpad provided on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0086] Those skilled in the art will understand that Figure 8 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include Figure 8 More or fewer components may be shown, or certain components may be combined, or may have a different arrangement of components.

[0087] In one embodiment, the present application provides Figure 7 The storage unit fault screening device shown can be implemented in the form of a computer program. The computer program can be used in Figure 8 The computer device shown in the figure is run. The memory of the computer device can store various program modules that constitute the storage unit fault screening device, such as the measurement startup module 510, the measurement preset value module 520, the remaining coulomb acquisition module 530, and the chip performance level acquisition module 540. The computer program composed of various program modules enables the processor to execute the steps of the storage unit fault screening method of each embodiment of the present application described in this specification. For example, in one embodiment, the computer executes step S10 through the measurement startup module 510. The computer device can execute step S20 through the measurement preset value module 520. And so on.

[0088] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the processor executes the steps of the storage unit fault screening method as described above.

[0089] In the embodiment of the present application, please refer to the above for the description of the above-mentioned storage unit fault screening method, which will not be repeated here.

[0090] In the embodiment of the present application, the program run based on the method stored in the storage medium of the embodiment of the present application can provide enterprises and server manufacturers with a reliable memory fault preprocessing solution, efficiently and accurately predict potential failures of memory particles in advance to ensure the stability and reliability of the server, guarantee the quality of server memory components after assembly, and achieve quantifiable improvement in server reliability indicators.

[0091] It should be understood that, although each step in the flow chart of each embodiment of the present application is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is clear explanation in this article, the execution of these steps does not have strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0092] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (RamBUS) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0093] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0094] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A storage unit fault screening circuit, wherein a plurality of storage units form a storage array, characterized in that: The circuit comprises: a control unit, a switch unit and a coulomb metering unit; The control unit is connected to the switch unit and is used to control the working state of each switch unit; The switch unit is connected to the memory to be tested and the coulomb metering unit, and a plurality of switch units are provided, each of which is connected to a corresponding storage array; the switch unit has two working states, working state 1 is to connect the coulomb metering unit to the storage array, and working state 2 is to connect the storage array to the ground line; The coulomb metering unit is used to measure the charge amount of the current introduced by the switch unit; The coulomb metering unit comprises: a coulomb metering chip, a measuring resistor and a grounding terminal; The first end of the measuring resistor is the input end of the coulomb metering unit, and is used to be connected to the switch unit, and the second end of the measuring resistor is grounded; The coulomb metering chip is used to measure the amount of charge flowing through the measuring resistor, and the coulomb metering chip has a positive measuring end and a negative measuring end, the positive measuring end is connected to the first end of the measuring resistor, and the negative measuring end is connected to the second end of the measuring resistor; The coulomb metering chip also has a measurement signal output terminal for outputting a coulomb measurement value.

2. A storage unit fault screening circuit according to claim 1, characterized in that: The switch unit comprises: a first input terminal, a first output terminal, a second output terminal and a control terminal; The first input terminal is used to connect to a storage array; The first output terminal is grounded; The second output terminal is connected to an input terminal of a coulomb metering unit; The control end is connected to a control signal output end of the control unit.

3. A storage unit fault screening circuit according to claim 2, characterized in that: There are four switch units, and the second output end of each switch unit is connected to a measurement bus, and the measurement bus is connected to the input end of the coulomb metering chip.

4. A storage unit fault screening method, characterized in that: The method is implemented based on the storage unit fault screening circuit according to any one of claims 1 to 3, and the method comprises the following steps: S10, connecting the measurement signal output terminal of the coulomb metering unit and the control signal input terminal of the control unit to the SMBUS bus of the memory to be tested; S20, turning off the memory interleaving function of the memory to be tested, parsing the address range of the storage array of the memory to be tested, and locating the data chip and storage array of the memory to be tested; each memory to be tested contains a number of storage blocks, each of the storage blocks contains a number of data chips, and each of the data chips contains a number of storage arrays; S30, supplying power to each storage array of the memory to be tested separately, and performing a write 1 operation; After a preset discharge time, the array is written to 0, and the remaining coulomb amount of the array after writing 0 is obtained based on the coulomb metering unit; S40, testing all the storage arrays of the memory to be tested one by one to obtain a residual coulomb set of all the storage arrays of each memory chip, and comparing the residual coulomb set with a preset performance parameter threshold to obtain a performance level of the memory chip.

5. A storage unit fault screening method according to claim 4, characterized in that: The method further comprises: S50, based on the performance parameter threshold, screening out faulty storage chips and faulty storage arrays that meet preset fault standards; S60, based on a memory fault test algorithm, accurately locate the fault interval in the faulty storage array to obtain an address of the fault interval; S70: Repair the fault interval based on a memory fault repair algorithm and the address of the fault interval.

6. A storage unit fault screening method according to claim 5, characterized in that: The method further comprises: S80. After completing step S70, repeat steps S10 to S40 for multiple times to obtain a memory chip whose remaining coulomb amount set is less than a preset value, and repair the memory chip based on steps S60 to S70 to improve the yield rate.

7. A storage unit fault screening device, characterized in that: The device is implemented based on the storage unit fault screening circuit according to any one of claims 1 to 3, and the storage unit fault screening device comprises: A measurement start module, used to connect the measurement signal output terminal of the coulomb metering unit and the control signal input terminal of the control unit to the SMBUS bus of the memory to be tested; The measurement preset value module is used to turn off the memory interleaving function of the memory to be tested, and parse the address range of the storage array of the memory to be tested, and locate the data chip and storage array of the memory to be tested; each memory to be tested contains a number of storage blocks, each of the storage blocks contains a number of data chips, and each of the data chips contains a number of storage arrays; The remaining coulomb acquisition module is used to supply power to each storage array of the memory to be tested separately and perform a write 1 operation; after a preset discharge time, the array is written 0, and the remaining coulomb amount of the array after writing 0 is obtained based on the coulomb metering unit; The chip performance level acquisition module is used to test all the storage arrays of the memory to be tested one by one, obtain the remaining coulomb set of all the storage arrays of each memory chip, compare the remaining coulomb set with the preset performance parameter threshold, and obtain the performance level of the memory chip.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor executes the steps of the storage unit fault screening method according to any one of claims 4 to 6.

9. A storage unit fault screening system, characterized in that: When the system is running, the steps of the storage unit fault screening method according to any one of claims 4 to 6 are executed.

Citation Information

Patent Citations

  • Storage array fault detection method and device and electronic equipment

    CN114550799A

  • Method and system for processing gate tube fault of memory array

    CN117831595A

  • NAND memory array incorporating multiple write pulse programming of individual memory cells and method for operation of same

    CN1910701A

  • Modular power pack energy storage unit

    US20230170732A1