Computation task execution method and device for mass spectrometer

Abstract

The invention relates to a calculation task execution method and device for a mass spectrometer. The method comprises the following steps: initializing an OpenCL platform; receiving a calculation task, and splitting the calculation task to obtain an extended message word; creating a corresponding kernel program through a CPU according to the number of the extended message words; and based on the OpenCL platform, executing the kernel program through an FPGA to perform parallel execution on the extended message word to obtain an execution result. According to the method, the calculation task of the mass spectrometer can be remarkably accelerated, so that higher throughput is realized when large-scale data is processed.

Description

Technical Field

[0001] The present invention relates to the technical field of data storage management, and in particular to a computing task execution method and device for a mass spectrometer. Background Art

[0002] During the mass spectrometry analysis, the mass spectrometer needs to analyze the intermediate processing information such as the mobile phase and ion flow, and the image spectrum peaks and text information formed in the end, so the data storage volume is huge and the relevant data is called in real time. This poses a challenge to the traditional mass spectrometer data calculation efficiency and storage system. The original CPU image calculation and hard disk storage are obviously no longer able to adapt to the development of today's mass spectrometry analysis technology.

[0003] At present, the traditional mass spectrometry calculation and storage strategy is too simple. The technical route is that the data is directly stored in the host's hard disk through relevant buses such as SATA bus, PCIE bus, etc. after being processed by the CPU. The disadvantages of this storage strategy are low calculation efficiency of image information, low efficiency of data storage and reading, and it is not suitable for the analysis of large amounts of data and the rapidly increasing and dynamically changing image information generated during the analysis process. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems existing in the related art. To this end, the present invention provides a method and device for executing a computing task for a mass spectrometer.

[0005] The present invention provides a method for executing a computing task for a mass spectrometer, comprising:

[0006] S1: Initialize the OpenCL platform;

[0007] S2: receiving a computing task, splitting the computing task, and obtaining an extended message word;

[0008] S3: creating a corresponding kernel program through the CPU according to the number of the extended message words;

[0009] S4: Based on the OpenCL platform, the kernel program is executed through the FPGA to execute the extended message words in parallel to obtain an execution result.

[0010] According to a computing task execution method for a mass spectrometer provided by the present invention, step S1 further comprises:

[0011] S11: query the OpenCL operating environment;

[0012] S12: query OpenCL device;

[0013] S13: Based on the OpenCL device, create an OpenCL object to complete the OpenCL platform initialization.

[0014] According to a computing task execution method for a mass spectrometer provided by the present invention, the OpenCL operating environment in step S11 includes an OpenCL version and an available OpenCL platform.

[0015] According to a computing task execution method for a mass spectrometer provided by the present invention, the OpenCL object in step S13 includes an OpenCL context, a command queue, a memory object, a program object, and a kernel object.

[0016] According to a method for executing a computing task for a mass spectrometer provided by the present invention, the expression of the splitting rule for splitting the computing task in step S2 is:

[0017] S n (x) = x>>n;

[0018] R n (x)=x< <n;

[0019]

[0020] W t =σ1(W t-2 )+W t-7 +σ0(W t-15 )+(W t-16 );

[0021] Where x is the input message word, n (n = 3, 7, 10, 17, 18, 19) is the number of bit-level shifts, S n (x) is the message word after x is shifted right by n bits, R n (x) is the message word after x is shifted left by n bits, σ0(x) is the first expansion coefficient, σ1(x) is the second expansion coefficient, t is the index value of the extended message word, W t is the calculated extended message word, W t-i (i=2,7,15,16) is W t The previous i-th message word.

[0022] According to a computing task execution method for a mass spectrometer provided by the present invention, step S3 further comprises:

[0023] S31: according to the number of the extended message words, establish kernel objects corresponding to the number of working groups;

[0024] S32: Fill the message extension word into the memory object;

[0025] S33: Submit the kernel object and the memory object to a command queue to wait for execution.

[0026] According to a computing task execution method for a mass spectrometer provided by the present invention, step S4 further comprises:

[0027] S41: executing the extended message word in parallel through the FPGA to obtain the intermediate variable stored in the local memory;

[0028] S42: Moving the intermediate variable from the local memory to the global memory to obtain a hash value stored in the GPU end;

[0029] S43: After all extended message words are executed, the hash value is transmitted to the CPU to obtain the execution result.

[0030] A second aspect of the present invention provides a computing task execution device for a mass spectrometer, comprising:

[0031] Initialization module: used to initialize the OpenCL platform;

[0032] Splitting module: used for receiving a computing task, splitting the computing task, and obtaining an extended message word;

[0033] A creation module: used for creating a corresponding kernel program through a CPU according to the number of the extended message words;

[0034] A parallel execution module is used to execute the kernel program through FPGA based on the OpenCL platform to execute the extended message word in parallel and obtain an execution result.

[0035] A third aspect of the present invention provides a computing task execution device for a mass spectrometer, comprising:

[0036] A memory and at least one processor, wherein instructions are stored in the memory;

[0037] The processor calls the instructions in the memory to enable the computing task execution device for a mass spectrometer to execute the computing task execution method for a mass spectrometer as described in any one of the above items.

[0038] A fourth aspect of the present invention provides a computer-readable storage medium having instructions stored thereon, and when the instructions are executed by a processor, the method for executing a computing task for a mass spectrometer as described in any one of the above items is implemented.

[0039] The present invention provides a method, device, equipment and storage medium for executing computing tasks of a mass spectrometer. By utilizing the parallel computing capabilities of the OpenCL platform and FPGA, the complex computing tasks generated by the mass spectrometer can be efficiently processed. The parallel processing characteristics provided enable multiple computing tasks to be performed simultaneously, thereby significantly improving the computing speed and efficiency. Secondly, the present invention dynamically creates and configures OpenCL kernel programs according to the complexity and scale of the computing tasks, and adjusts the number of FPGA workgroups, thereby achieving flexible allocation and efficient use of computing resources, so that the present invention can adapt to mass spectrometer computing tasks of different scales and complexities, and has good scalability. In addition, the present invention realizes flexible allocation and efficient use of computing resources by fine task splitting and internal The memory management effectively utilizes computing resources and reduces resource waste. At the same time, the context and command queue management of the OpenCL platform further improves the efficiency and accuracy of resource scheduling. Secondly, the present invention ensures the accuracy and reliability of computing tasks through specific splitting rules and calculation processes. In particular, when processing complex data generated by a mass spectrometer, the accuracy and reliability of data processing can be significantly improved through precise calculation and memory management. Finally, compared with traditional CPU or GPU-based computing methods, the present invention generally has lower energy consumption by using FPGA for parallel computing. This enables the present invention to achieve a significant improvement in energy efficiency while maintaining high performance, which is beneficial to reducing the overall operating cost of the mass spectrometer.

[0040] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the present invention or 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 some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1 It is a flow chart of a method for executing a computing task for a mass spectrometer provided by the present invention;

[0043] Figure 2 It is a schematic diagram of the structure of a computing task execution system for a mass spectrometer provided by the present invention; Figure 3 Schematic diagram of the performance comparison between the reconfigurable hash algorithm designed for the present invention and the traditional hash function.

[0044] Reference numerals:

[0045] 100, initialization module; 200, split module; 300, creation module; 400, parallel execution module. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical scheme and advantages of the present invention clearer, the technical scheme of the present invention will be clearly and completely described below in conjunction with the drawings in the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present invention. The following embodiments are used to illustrate the present invention, but cannot be used to limit the scope of the present invention.

[0047] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the embodiment of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0048] In order to better understand the present invention, the application and background of the present invention are first explained in detail below.

[0049] The triple quadrupole mass spectrometer is widely used in the fields of reconnaissance and environmental science as a conventional quantitative analysis instrument due to its high sensitivity, fast analysis speed and small sample consumption. When using a triple quadrupole mass spectrometer, the mobile phase sample is first placed in the injection system. After the sample is pre-treated by the injection system, it enters the ion source and the triple quadrupole vacuum chamber in turn, so that the mobile phase is converted into an ion flow signal. Finally, the ion flow signal passes through the signal processing circuit to generate the image spectrum peak information and text information to be obtained by mass spectrometry analysis.

[0050] During the mass spectrometry analysis process, it is necessary to simultaneously analyze the intermediate processing information such as the mobile phase and ion flow, plus the final image spectrum peaks and text information. The data storage volume is extremely large, and the relevant data is called in real time. This poses a challenge to the data calculation efficiency and storage system of the traditional triple quadrupole mass spectrometer. The original CPU image calculation and hard disk storage are obviously no longer able to adapt to the development of today's mass spectrometry analysis technology.

[0051] Today, traditional mass spectrometry calculation and storage strategies are too simplistic. The technical route is that after the data is processed by the CPU, it is directly stored in the host's hard disk through related buses, such as SATA bus, PCIE bus, etc. The disadvantages of this are: low efficiency in calculating image information, low efficiency in data storage and reading, and it is not suitable for the analysis of large amounts of data and the rapidly increasing and dynamically changing image information generated during the analysis process.

[0052] With the development of computer heterogeneous technology and computer vision technology, the rapid classification and accelerated parallel computing of heterogeneous processing architectures provide effective support for dynamic real-time processing of images. Based on this, the present invention provides a computing task execution method, device, equipment and storage medium for a mass spectrometer, which mainly designs a heterogeneous computing unit of CPU+multi-core GPU+FPGA, uses the characteristics of the CPU to calculate related instructions, and for high-intensity image computing tasks of single-precision floating points, assigns them to multi-core GPU processors for image computing, and implements a reconfigurable hash algorithm based on the heterogeneous computing standard OpenCL in the FPGA.

[0053] In order to better understand the present invention, the terms appearing in the embodiments of the present invention are explained in detail below.

[0054] Open Computer Language (OpenCL) is a standard for heterogeneous computing. It aims to enable computing devices of different specifications to achieve collaborative computing through standardized programming models to achieve higher performance. It is widely used in embedded development, image processing, high-performance computing and other fields. OpenCL supports a variety of computing devices, such as CPU, GPU, FPGA, etc. OpenCL divides computing devices into host and device, and the data communication between the two is generally through PCI-Express communication. The application of OpenCL consists of a host program and one or more kernels, which run on the host and device respectively.

[0055] Hash algorithms are also generally called hash algorithms or message digest algorithms because of the calculation characteristics of compression functions and the fact that different hash values ​​are generated for different input data. Hash algorithms are widely used in the field of information security due to the above characteristics. The design proposed by the present invention improves the flexibility of system calculation and the efficiency of image calculation of triple quadrupole mass spectrometer.

[0056] Combine the following Figures 1 to 3 Embodiments of the present invention are described.

[0057] like Figure 1 As shown, the present invention provides a computing task execution method for a mass spectrometer, comprising:

[0058] S1: Initialize the OpenCL platform.

[0059] Furthermore, the purpose of step S1 is to initialize the OpenCL computing platform by the CPU program, specifically by first querying the OpenCL operating environment and OpenCL devices, and creating various OpenCL objects to establish a connection between the CPU and the FPGA. The initialization operation of the OpenCL computing platform only needs to be executed once at the beginning, and does not need to be run again when the accelerator accelerates different hash algorithms.

[0060] Wherein, step S1 further comprises:

[0061] S11: Query the OpenCL operating environment.

[0062] The OpenCL operating environment in step S11 includes an OpenCL version and an available OpenCL platform.

[0063] S12: Query the OpenCL device.

[0064] In steps S11 to S12, the CPU program is first started and executed to prepare to initialize the OpenCL computing platform. Then, the CPU program queries the OpenCL operating environment on the system, including the supported OpenCL version, available OpenCL platforms, etc., and then queries the OpenCL device. The OpenCL device can be a CPU, GPU or FPGA. In this embodiment, the focus is on FPGA devices.

[0065] S13: Based on the OpenCL device, create an OpenCL object to complete the OpenCL platform initialization.

[0066] The OpenCL objects in step S13 include an OpenCL context, a command queue, a memory object, a program object, and a kernel object.

[0067] The purpose of step S13 is to create an OpenCL object, in which the context is used to manage the relationship between OpenCL objects, the command queue is used to submit and execute commands between the CPU and FPGA, the memory object is used to transfer data between the CPU and FPGA, the program object is used to load and compile the OpenCL program (kernel code), and the kernel object is used to perform specific computing tasks on the FPGA.

[0068] S2: receiving a computing task, splitting the computing task, and obtaining an extended message word.

[0069] The expression of the splitting rule for splitting the computing task in step S2 is:

[0070] S n (x) = x>>n;

[0071] R n (x)=x< <n;

[0072]

[0073] W t =σ1(W t-2 )+W t-7 +σ0(W t-15 )+(W t-16 );

[0074] Where x is the input message word, n (n = 3, 7, 10, 17, 18, 19) is the number of bit-level shifts, S n (x) is the message word after x is shifted right by n bits, R n (x) is the message word after x is shifted left by n bits, σ0(x) is the first expansion coefficient, σ1(x) is the second expansion coefficient, t is the index value of the extended message word, W t is the calculated extended message word, W t-i (i=2,7,15,16) is W t The previous i-th message word.

[0075] The core part of the hash algorithm is the calculation of the compression function f. In step S2, the present invention designs a reconfigurable hash function, the number of iterations of which is determined by FPGA according to different image information classifications, and is therefore defined as reconfigurable.

[0076] The reconfigurable hash function provided by the present invention is different from the traditional hash function in that, for example, the length of the input CV of the compression function f is 256 bits, which can be split into 8 32-bit data. If the length of the packet data is 512 bits, it is split into 16 message words W0, W1, W2, ..., W15, and more message words W16, W17, W18, ..., W63 can be calculated by the above formula.

[0077] In the implementation process of the specific step S2, first, a basic hash algorithm is selected according to the computing task as the basic framework of the reconfigurable hash function; then the CPU side preprocesses and analyzes the input image information to determine its classification; secondly, based on the classification result of the image information, the number of iterations that the reconfigurable hash function should execute is determined; finally, the determined number of iterations and other necessary configuration parameters are passed to the reconfigurable hash function on the FPGA.

[0078] S3: Creating a corresponding kernel program through the CPU according to the number of the extended message words.

[0079] Wherein, step S3 further comprises:

[0080] S31: According to the number of the extended message words, establish kernel objects corresponding to the number of working groups.

[0081] S32: Fill the message extension word into the memory object.

[0082] S33: Submit the kernel object and the memory object to a command queue to wait for execution.

[0083] In steps S31 to S33, first, according to the selected hash algorithm, a corresponding kernel object is created from the compiled OpenCL program, and then the workgroup size is set for the kernel object, that is, the number of kernel instances executed in parallel, so as to optimize the computing performance on the FPGA. After the setting, the CPU program prepares the input data and fills the data into the OpenCL memory object. Finally, the kernel object and related parameters, including input data and workgroup size, are submitted to the command queue, and the kernel program is executed by the FPGA.

[0084] S4: Based on the OpenCL platform, the kernel program is executed through the FPGA to execute the extended message words in parallel to obtain an execution result.

[0085] Wherein, step S4 further comprises:

[0086] S41: Execute the extended message word in parallel through the FPGA to obtain the intermediate variables stored in the local memory.

[0087] S42: The intermediate variable is moved from the local memory to the global memory to obtain a hash value stored in the GPU.

[0088] S43: After all extended message words are executed, the hash value is transmitted to the CPU to obtain the execution result.

[0089] After steps S1 to S3 are prepared, in step S4, the message word expansion and iterative calculation of the hash algorithm are executed on the FPGA in a full pipeline structure, and the intermediate variables of the calculation are also stored in the local memory.

[0090] When executed, the hash function can map the input of the infinite domain to the hash value of the finite domain. The compression mapping relationship is realized by the compression function. In this embodiment, the Merkle-Damgard structure is adopted, and its expression is:

[0091] CV0 = IV;

[0092] CVi =f(CV i-1 ,M i-1 );

[0093] H(M)=CV L ;

[0094] Among them, H is the hash function, M is the input data of the hash, CV is the link variable IV is the initial vector, f is the compression function, L is the total number of grouped data, and M can be divided into multiple grouped data M0, M1, M2, ..., M L-1 , respectively correspond to L rounds of iterative calculations of the compression function f, and the input data for the i-th round of compression function calculation is the grouped data M i-1 And the calculation result CV of the last compression function i The hash function sets the initial vector IV as CV0 for the first compression function calculation. When the compression function calculates all the grouped data, the final calculation result CV L This is the final hash value.

[0095] Among them, the hash value, as the calculation result, is first moved from the local memory to the global memory and waits to be transmitted to the GPU side. When all calculation tasks are completed, all hash values ​​are transmitted to the CPU through PCIe.

[0096] like Figure 2 As shown, the present invention also provides a computing task execution device for a mass spectrometer, comprising:

[0097] Initialization module 100: used to initialize the OpenCL platform;

[0098] Splitting module 200: used for receiving a computing task, splitting the computing task, and obtaining an extended message word;

[0099] Creation module 300: used to create a corresponding kernel program through the CPU according to the number of the extended message words;

[0100] The parallel execution module 400 is used to execute the kernel program through the FPGA based on the OpenCL platform to execute the extended message word in parallel and obtain the execution result.

[0101] The present invention also provides a computing task execution device for a mass spectrometer, comprising:

[0102] A memory and at least one processor, wherein instructions are stored in the memory;

[0103] The processor calls the instructions in the memory to enable the computing task execution device for a mass spectrometer to execute the computing task execution method for a mass spectrometer as described in any one of the above items.

[0104] The present invention also provides a computer-readable storage medium having instructions stored thereon, and when the instructions are executed by a processor, the method for executing a computing task for a mass spectrometer as described in any one of the above items is implemented.

[0105] Furthermore, a computing task execution device for a mass spectrometer provided by the present invention may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU), for example, one or more processors and memories, one or more storage media for storing applications or data, such as one or more mass storage devices, wherein the memories and storage media may be short-term storage or permanent storage, and the program stored in the storage medium may include one or more modules, each module may include a series of instruction operations in the computing task execution device for a mass spectrometer, and further, the processor may be configured to communicate with the storage medium to execute a series of instruction operations in the storage medium on the computing task execution device for a mass spectrometer.

[0106] It may also include one or more power supplies, one or more wired or wireless network interfaces, one or more input and output interfaces, and one or more operating systems, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will appreciate that the computing task execution device structure for a mass spectrometer provided by the present invention does not constitute a limitation on the device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0107] In a specific embodiment, a computing unit formed by the computing task execution method, system and storage medium for a mass spectrometer provided by the present invention is composed of a CPU+multi-core GPU+FPGA. The CPU+multi-core GPU uses HiSilicon's hi3798V200 high-performance processor, and the FPGA uses Kintex-7. The interaction between the heterogeneous device CPU and FPGA is realized through the OpenCL protocol and PCIE bus. OpenCL defines a variety of abstract memory models, divides a multi-level storage structure according to memory access speed and storage capacity, and implements it in the device with different hardware resources.

[0108] Figure 3 The performance comparison diagram of the reconfigurable hash algorithm designed for the present invention and the traditional hash functions SHA-1 and RIPEMD-160 is shown in FIG. Figure 3As shown in the figure, it can be seen that the computing performance of the reconfigurable hash algorithm provided by the present invention is significantly higher than other traditional hash functions. Therefore, the application of the present invention significantly improves the processing efficiency and storage efficiency of image information in the triple quadrupole mass spectrometer.

[0109] The present invention provides a computing task execution method, system and storage medium for a mass spectrometer, which are mainly divided into heterogeneous hardware accelerated computing and a reconfigurable hash algorithm. The hardware accelerated computing part is used to improve the computing efficiency of image information, and the reconfigurable hash algorithm is mainly used for image compression, thereby increasing data storage space.

[0110] The present invention can significantly accelerate the computing tasks of the mass spectrometer by utilizing the parallel computing capabilities of the OpenCL platform and FPGA. The parallelism of the FPGA enables multiple computing units to work simultaneously, thereby achieving higher throughput when processing large-scale data. This efficient parallel computing capability is crucial for scientific instruments such as mass spectrometers that need to process large amounts of complex data, and can greatly shorten data processing time and improve experimental efficiency.

[0111] After receiving a computing task, the present invention will split it according to the complexity and scale of the task and generate extended message words, which not only helps to decompose large tasks into subtasks that are easier to manage, but also makes resource allocation more flexible. According to the number of extended message words, the CPU can dynamically create and configure corresponding kernel programs to ensure that computing resources are fully utilized. The flexibility enables the present invention to adapt to mass spectrometer computing tasks of different scales and types, thereby improving the adaptability and scalability of the system.

[0112] The present invention achieves optimal utilization of computing resources through sophisticated task management and resource scheduling. The hardware acceleration feature of FPGA enables computing tasks to be completed with lower energy consumption and higher efficiency, which helps to reduce operating costs compared to traditional CPU or GPU solutions. At the same time, the standardized interface of the OpenCL platform enhances the compatibility between different hardware, further reducing the complexity of system integration and maintenance.

[0113] The present invention ensures the accuracy and reliability of the calculation task by adopting precise splitting rules and calculation processes. In particular, when processing complex data, the parallel calculation and precise control of FPGA can significantly reduce the calculation error and improve the accuracy and reliability of data processing.

[0114] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. 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 method for executing a computing task for a mass spectrometer, characterized in that: include: S1: Initialize the OpenCL platform; S2: receiving a computing task, splitting the computing task, and obtaining an extended message word; S3: creating a corresponding kernel program through the CPU according to the number of the extended message words; S4: Based on the OpenCL platform, the kernel program is executed through the FPGA to execute the extended message words in parallel to obtain an execution result.

2. A method for executing a computing task for a mass spectrometer according to claim 1, characterized in that: Step S1 further comprises: S11: query the OpenCL operating environment; S12: query OpenCL device; S13: Based on the OpenCL device, create an OpenCL object to complete the OpenCL platform initialization.

3. A method for executing a computing task for a mass spectrometer according to claim 2, characterized in that: The OpenCL operating environment in step S11 includes an OpenCL version and an available OpenCL platform.

4. A method for executing a computing task for a mass spectrometer according to claim 2, characterized in that: The OpenCL objects in step S13 include an OpenCL context, a command queue, a memory object, a program object, and a kernel object.

5. The method for executing a computing task for a mass spectrometer according to claim 1, characterized in that: The expression of the splitting rule for splitting the computing task in step S2 is: S n (x)=x>>n; R n (x)=x<<n; IN t =σ1(W t-2 )+W t-7 +σ0(W t-15 )+(In t-16 ); Where x is the input message word, n (n = 3, 7, 10, 17, 18, 19) is the number of bit-level shifts, S n (x) is the message word after x is shifted right by n bits, R n (x) is the message word after x is shifted left by n bits, σ0(x) is the first expansion coefficient, σ1(x) is the second expansion coefficient, t is the index value of the extended message word, W t is the calculated extended message word, W t-i (i=2,7,15,16) is W t The previous i-th message word.

6. A method for executing a computing task for a mass spectrometer according to claim 1, characterized in that: Step S3 further comprises: S31: according to the number of the extended message words, establish kernel objects corresponding to the number of working groups; S32: Fill the message extension word into the memory object; S33: Submit the kernel object and the memory object to a command queue to wait for execution.

7. A method for executing computing tasks for a mass spectrometer according to claim 1, characterized in that: Step S4 further comprises: S41: executing the extended message word in parallel through the FPGA to obtain the intermediate variable stored in the local memory; S42: Moving the intermediate variable from the local memory to the global memory to obtain a hash value stored in the GPU end; S43: After all extended message words are executed, the hash value is transmitted to the CPU to obtain the execution result.

8. A computing task execution device for a mass spectrometer, characterized in that: include: Initialization module: used to initialize the OpenCL platform; Splitting module: used for receiving a computing task, splitting the computing task, and obtaining an extended message word; A creation module: used for creating a corresponding kernel program through a CPU according to the number of the extended message words; A parallel execution module is used to execute the kernel program through FPGA based on the OpenCL platform to execute the extended message word in parallel and obtain an execution result.

9. A computing task execution device for a mass spectrometer, characterized in that: include: A memory and at least one processor, wherein instructions are stored in the memory; The processor calls the instructions in the memory to enable the computing task execution device for a mass spectrometer to execute the computing task execution method for a mass spectrometer according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed by the processor, the computing task execution method for a mass spectrometer according to any one of claims 1 to 7 is implemented.

Images