CN113241120A - Gene sequencing system and sequencing method - Google Patents
Gene sequencing system and sequencing method Download PDFInfo
- Publication number
- CN113241120A CN113241120A CN202110632394.2A CN202110632394A CN113241120A CN 113241120 A CN113241120 A CN 113241120A CN 202110632394 A CN202110632394 A CN 202110632394A CN 113241120 A CN113241120 A CN 113241120A
- Authority
- CN
- China
- Prior art keywords
- management
- data
- sequencing
- gene
- memory
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012163 sequencing technique Methods 0.000 title claims abstract description 261
- 108090000623 proteins and genes Proteins 0.000 title claims abstract description 231
- 238000000034 method Methods 0.000 claims abstract description 137
- 238000012545 processing Methods 0.000 claims abstract description 126
- 230000008569 process Effects 0.000 claims abstract description 119
- 238000004364 calculation method Methods 0.000 claims description 68
- 230000006870 function Effects 0.000 claims description 17
- 238000004886 process control Methods 0.000 claims description 8
- 238000012216 screening Methods 0.000 claims description 5
- 238000007726 management method Methods 0.000 description 205
- 238000007906 compression Methods 0.000 description 17
- 238000004458 analytical method Methods 0.000 description 15
- 230000006837 decompression Effects 0.000 description 14
- 230000006835 compression Effects 0.000 description 12
- 238000004891 communication Methods 0.000 description 8
- 230000003993 interaction Effects 0.000 description 7
- 230000005540 biological transmission Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 238000004148 unit process Methods 0.000 description 4
- 102000057361 Pseudogenes Human genes 0.000 description 3
- 108091008109 Pseudogenes Proteins 0.000 description 3
- 230000001133 acceleration Effects 0.000 description 3
- 238000013144 data compression Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000002068 genetic effect Effects 0.000 description 3
- 238000013439 planning Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 239000012634 fragment Substances 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 238000003766 bioinformatics method Methods 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 230000003139 buffering effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012252 genetic analysis Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 102000039446 nucleic acids Human genes 0.000 description 1
- 108020004707 nucleic acids Proteins 0.000 description 1
- 150000007523 nucleic acids Chemical class 0.000 description 1
- 239000002773 nucleotide Substances 0.000 description 1
- 125000003729 nucleotide group Chemical group 0.000 description 1
- 102000004169 proteins and genes Human genes 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
- G16B30/00—ICT specially adapted for sequence analysis involving nucleotides or amino acids
Landscapes
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biophysics (AREA)
- Proteomics, Peptides & Aminoacids (AREA)
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Biotechnology (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Theoretical Computer Science (AREA)
- Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)
Abstract
The invention relates to the technical field of gene sequencing data processing, in particular to a gene sequencing system and a gene sequencing method. By adopting the scheme, the processing steps in the gene sequencing process are unloaded to the central processing unit, the memory computing unit and the storage computing unit, so that the parallel processing of gene data is realized, the parallel efficiency among different processing steps of gene sequencing is effectively improved, and the performance of the gene sequencing processing process is improved.
Description
Technical Field
The invention relates to the technical field of gene sequencing data processing, in particular to a gene sequencing system and a gene sequencing method.
Background
With the rapid development of bioinformatics, genetic analysis has become a widely used technical means in scientific research and industry, and has been successfully applied to species identification, disease diagnosis, and the like, wherein genetic sequencing has become an increasingly important field in genetic research, and generally, genetic sequencing relates to determining the nucleotide sequence of nucleic acids such as RNA or DNA fragments. By analyzing shorter gene sequences, the resulting sequence information is used in various bioinformatics methods to logically fit multiple fragments together to reliably determine sequences of a broader length of genetic material.
The gene sequencing technology is closely related to the computer technology, and the whole computer processing flow of the gene sequencing can be roughly divided into six steps: BWA-MEM, Sort, Mark Duplicate, Indel reading, BQSR, and variable Calling. The existing gene sequencing process is usually controlled by a CPU, and meanwhile, corresponding computer processing processes are all carried out in the CPU, so that the load of the CPU is large, the processing performance is low, and the single gene sequencing process needs to spend much time.
Therefore, a gene sequencing system and a gene sequencing method capable of improving the processing performance of the gene sequencing process are needed.
Disclosure of Invention
One of the objectives of the present invention is to provide a gene sequencing system, which can improve the performance of the gene sequencing processing flow.
The invention provides a basic scheme I: a gene sequencing system comprises a central processing unit, an in-memory computing unit and a storage computing unit which can run in parallel at the same time, wherein the central processing unit is used for calling and executing a central sequencing operator, the in-memory computing unit is used for calling and executing an in-memory sequencing operator, and the storage computing unit is used for calling and executing a storage sequencing operator.
The beneficial effects of the first basic scheme are as follows: the central processing unit is arranged to call and execute a central sequencing operator to process the gene data; the setting of the memory computing unit calls and executes a memory sequencing operator to process the gene data; and the storage computing unit is used for calling and executing a storage sequencing operator to process the gene data, wherein the central sequencing operator, the in-memory sequencing operator and the storage sequencing operator are operators for performing different processing on the gene data in the gene sequencing process.
The method comprises the steps of distributing different processing steps in a gene sequencing process in different units, simultaneously operating a central processing unit, a memory computing unit and a storage computing unit in parallel, wherein the gene sequencing process comprises a step A, a step B and a step C, respectively operating the three steps in the central processing unit, the memory computing unit and the storage computing unit, and when the memory computing unit processes the step B in the current gene sequencing process, the central processing unit processes the step A in another gene sequencing process in parallel. Through the parallel processing of a plurality of units, the parallel efficiency among different processing steps of gene sequencing is effectively improved.
By adopting the scheme, the processing steps in the gene sequencing process are unloaded to different units to run in parallel, so that the parallel processing of gene data is realized, the parallel efficiency among different processing steps of gene sequencing is effectively improved, and the performance of the gene sequencing processing process is improved.
Further, an in-memory computing unit comprising:
the main memory module is used for caching gene data in a gene sequencing process;
and the logic calculation module is used for calling the gene data in the main memory module, calling a preset in-memory sequencing operator and executing the in-memory sequencing operator to process the called gene data.
Has the advantages that: and the main memory module is used for storing intermediate data used in the gene sequencing process, and the intermediate data is processed gene data in the gene sequencing process. And the logic calculation module is arranged to call and execute the in-memory sequencing operator to process the gene data, and the processed gene data is the data called from the main memory module. The main memory module and the logic calculation module are arranged in the memory calculation unit together, gene data required by the logic calculation module for processing are stored in the main memory module, the physical distance between the main memory module and the logic calculation module is shortened, the data moving expense is reduced, and the data processing performance of the memory calculation unit is improved. Meanwhile, certain steps in the gene sequencing process are realized by adopting a special logic calculation module, compared with the traditional software, the processing speed is higher, and the data processing performance in the gene sequencing process is further improved.
Furthermore, the central processing unit comprises a central control module and a customized operation function module which can run in parallel at the same time, the customized operation function module is used for calling and executing a central sequencing operator, and the central processing module is used for calling and executing other sequencing operators except the central sequencing operator, the in-memory sequencing operator and the storage sequencing operator.
Has the advantages that: the customized operation functional module is special hardware and is used for calling and executing a central sequencing operator; the central processing module may be considered a CPU for invoking and executing other sequencing operators. The data volume and the required computing resource amount of the gene sequencing are considered, certain steps in the gene sequencing process are still kept in the central processing module for carrying out, the customized operation function module and the central control module are run in parallel in the central processing unit, and the parallel processing efficiency of each processing step of the gene sequencing is ensured.
Further, the memory computing unit and the storage computing unit both comprise a flow control module;
the flow control module is used for receiving the flow management information, arbitrating the flow management information to obtain a management command, and distributing the management command to a preset management message queue; the management device is also used for receiving the data information, calling the management command in the corresponding management message queue according to the data information and executing the management command;
the memory computing unit and the storage computing unit are also used for calling and executing the corresponding memory sequencing operator or storage sequencing operator when executing the management command.
Has the advantages that: the flow control module is used for managing the gene sequencing processes in the memory computing unit and the memory computing unit, and the flow management information is used for managing programmability, so that the centralized management of the gene sequencing processes is realized.
The process management information comprises a plurality of management commands, the process control module arbitrates the process management information to obtain the plurality of management commands, and the management commands are distributed to the management message queue for storage. And when the data information is received, calling the management command from the corresponding management message queue for execution, thereby realizing the processing and transmission of the data information. Meanwhile, the processing and transmission of the data information are unloaded to a plurality of units for processing, so that the delay of data information control is reduced, and the efficient control of the gene sequencing process is realized.
By adopting the scheme, programmable gene sequencing flow control is provided through the flow control module and the flow management information, and meanwhile, the centralized management of the gene sequencing flow is realized, so that the time overhead of the system in the gene sequencing flow control is effectively reduced, and the performance of the gene sequencing processing flow is further improved.
Furthermore, the management message queue comprises a queue number, the process control module comprises a management message arbitration submodule, and the management message arbitration submodule is used for analyzing the process management information to obtain the message queue number and writing the management command into the corresponding management message queue in sequence according to the message queue number and the queue number.
Has the advantages that: the management message queue in the flow control module has a unique queue number, and the management message arbitration submodule is arranged to obtain the message queue number from the flow management information so as to obtain the management message queue to which the management command needs to be put. The operation sequence executed in the gene sequencing process is fixed, so that the management commands are written into the management message queue in sequence, and the subsequent management commands can be called quickly.
Further, the management message queue comprises a queue number, the process control module comprises a data flow submodule and a multi-queue submodule, the data flow submodule is used for analyzing data information to obtain the data queue number, the corresponding management message queue is screened according to the data queue number and the queue number, a management command in the screened management message queue is called, and the multi-queue submodule is used for deleting the corresponding management command in the management message queue when the management command in the management message queue is called.
Has the advantages that: the data information includes gene data and a queue number of a management message queue in which a gene sequencing process executed by the gene data is located, that is, a data queue number. And the data flow sub-module is arranged, a corresponding management message queue is screened based on the data queue number and the queue number, and a management command is called to execute, so that the gene sequencing process is completed. And the multi-queue submodule is arranged to delete the corresponding management command after the management command is called, so that the command executed in the next gene sequencing process is positioned at the head of the management message queue, and the management command is quickly called when the next data information comes.
Another object of the present invention is to provide a method for gene sequencing.
The invention provides a second basic scheme: a method of gene sequencing comprising:
a central processing step: calling and executing a preset central sequencing operator through a central processing unit;
an in-memory calculation step: calling and executing a preset in-memory sequencing operator through an in-memory computing unit;
and a storage calculation step: calling and executing a preset storage sequencing operator through a storage computing unit;
the central processing step, the in-memory computing step and the storage computing step may be run concurrently and in parallel.
The second basic scheme has the beneficial effects that: the gene data is processed by the central processing unit, the memory computing unit and the storage computing unit. The method comprises the steps of a central processing step, an in-memory calculation step, a storage calculation step and a storage sequencing step, wherein the central processing step calls and executes a central sequencing operator to process gene data, the in-memory calculation step calls and executes an in-memory sequencing operator to process the gene data, and the storage calculation step calls and executes a storage sequencing operator to process the gene data, wherein the central sequencing operator, the in-memory sequencing operator and the storage sequencing operator are operators for performing different processing on the gene data in a gene sequencing process.
By adopting the scheme, different processing steps in the gene sequencing process are distributed in different units, so that parallel processing of gene data is realized, the parallel efficiency among different processing steps of gene sequencing is effectively improved, and the performance of the gene sequencing process is improved.
Further, the memory computing step and the storage computing step include the following:
storing gene data in a gene sequencing process;
and calling the stored gene data, calling a preset in-memory sequencing operator or a storage sequencing operator, and executing the in-memory sequencing operator or the storage sequencing operator to process the called gene data.
Has the advantages that: and the physical distance between data storage and data processing is shortened, the data moving expense is reduced, the efficient calculation of gene data is realized, and the data processing performance of gene sequencing is improved.
Further, the following contents are included:
command management step: receiving flow management information, arbitrating the flow management information to obtain a management command, and distributing the management command to a preset management message queue;
a command execution step: and receiving the data information, calling a management command in the management message queue according to the data information, and calling an in-memory calculation step or a storage calculation step to execute the management command.
Has the advantages that: the process management information comprises a plurality of management commands, the process management information is arbitrated to obtain the plurality of management commands, and the management commands are distributed to the management message queue for storage. And when the data information is received, calling the management command from the corresponding management message queue and executing the corresponding memory calculation step and the storage calculation step, thereby realizing the processing and transmission of the data information.
By adopting the scheme, programmable gene sequencing flow control is provided through flow management information, centralized management of the gene sequencing flow is realized, and time overhead of the system in the gene sequencing flow control is effectively reduced.
Further, the management message queue includes a queue number, and distributes the management command to a preset management message queue, including the following:
analyzing the flow management information to obtain a message queue number, and writing the management commands into the corresponding management message queues in sequence according to the message queue number and the queue number;
invoking a management command in the management message queue according to the data information, wherein the management command comprises the following contents:
analyzing the data information to obtain a data queue number, screening a corresponding management message queue according to the data queue number and the queue number, calling a management command in the screened management message queue, and deleting the corresponding management command in the management message queue.
Has the advantages that: the management message queue has a unique queue number, and the message queue number is obtained by analyzing the flow management information, so that the management message queue to which the management command needs to be put is obtained. The operation sequence executed in the gene sequencing process is fixed, so that the management commands are written into the management message queue in sequence, the subsequent management commands can be called quickly, and the efficient control of the gene sequencing process is realized.
The data information includes gene data and a queue number of a management message queue in which a gene sequencing process executed by the gene data is located, that is, a data queue number. And screening the corresponding management message queue based on the data queue number and the queue number, and calling a management command to execute so as to complete the gene sequencing process. And after the management command is called, deleting the corresponding management command, so that the command executed in the next gene sequencing process is positioned at the first position of the management message queue, realizing the quick calling of the management command when the next data information comes, and reducing the time overhead of the system in the control of the gene sequencing process.
Drawings
FIG. 1 is a logic diagram of a gene sequencing system according to the present invention;
FIG. 2 is a logic diagram of a custom arithmetic function module in a gene sequencing system according to the present invention;
FIG. 3 is a logic diagram of an instruction parsing module in a gene sequencing system according to the present invention;
FIG. 4 is a logic diagram of a accelerator array module in a gene sequencing system according to the present invention;
FIG. 5 is a schematic diagram of a dynamic programming operator in a gene sequencing system according to the present invention;
FIG. 6 is a schematic diagram of the structure of a flow control module in a gene sequencing system according to the present invention.
Detailed Description
The following is further detailed by way of specific embodiments:
example one
A gene sequencing system, as shown in FIG. 1, comprises a central processing unit, a memory computing unit and a storage computing unit which can run simultaneously and in parallel, and further comprises a root complex.
The root complex is used for respectively communicating with the central processing unit, the memory computing unit and the storage computing unit, and the communication comprises data transmission of gene data and interaction of management messages. Specifically, the root complex is connected to the central processing unit via the FSB, the root complex is connected to the memory computing unit via the DIMM, and the root complex is connected to the memory computing unit via the PCIe.
The central processing unit is connected with the root complex through the FSB, sends control and management messages to the memory computing unit and the storage computing unit, receives completion events, sends and receives gene data, and achieves management and communication of the memory computing unit and the storage computing unit. The root complex is connected with the memory computing unit through the DIMM, performs interaction of data and management messages with the memory computing unit, and is connected with the memory computing unit through the PCIe, so that the memory computing unit and the memory computing unit can mutually send and receive DMA data and send and receive control messages. The central processing unit, the memory computing unit and the storage computing unit are managed in a centralized way through the root complex, and mutual communication among the units is realized.
The central processing unit is used for calling and executing a preset central sequencing operator, and the central sequencing operator comprises a dynamic programming sequencing operator and a hidden Markov model sequencing operator. Specifically, the central processing unit comprises a customized operation function module, the customized operation function module is used for calling and executing a central sequencing operator, and in the gene sequencing process, the gene data are processed according to a dynamic programming sequencing operator and a hidden Markov model sequencing operator. Specifically, the customized operation function module adopts a special hardware structure, and the Smith-Waterman dynamic algorithm used in the BWA-MEM and Indel reading steps in the gene sequencing process is unloaded to the customized operation function module, and in this embodiment, the customized operation function module adopts a pulse array.
In other embodiments, the central processing unit is further configured to manage a gene sequencing process, and further includes a central control module that can run concurrently with the custom operation function module, where the central control module calls and executes other sequencing operators except for the central sequencing operator, the in-memory sequencing operator, and the storage sequencing operator, and the other sequencing operators are sequencing operators used in other steps in the gene sequencing process except for steps executed by the custom operation function module, the in-memory computing unit, and the storage computing unit. The central control module is also used for managing a gene sequencing process and a programming interface of gene sequencing. The central control module adopts a general processor, and the whole gene sequencing process is managed by programming in the general processor.
As shown in fig. 2, the customized operation function module includes a general computation module, an instruction parsing module, an accelerator array module, and a configuration storage module.
The general computation module is a control core of the whole system, is used for executing the algorithm part executed by the non-accelerator array module, and is also used for splitting an instruction set of the dynamic programming algorithm and distributing instruction information required to be executed by the accelerator array module to the instruction analysis module. In this embodiment, the instruction set adopts RISC-V instruction set (reduced instruction set).
The instruction parsing module is used for providing interaction for the general computing module and the accelerator array module. In order to accomplish efficient interaction, in this embodiment, the instruction parsing module is further configured to receive instruction information in parallel, parse the instruction information in parallel, and the like.
The accelerator array module is a core acceleration component of the system, and is configured to complete acceleration of a specific algorithm according to a result of the analysis, where the specific algorithm is a dynamic programming algorithm in this embodiment. In order to achieve high-performance and high-throughput calculation, in this embodiment, the accelerator array module further has functions of parallel instruction information receiving, accelerator module arbitration, dynamic programming algorithm acceleration, calculation granularity support, and the like.
The configuration storage module is used for providing calculated data and parameters of dynamic planning calculation for the general calculation module and the accelerator array module, and the functions of data storage and data supply are completed. The calculated data includes reference sequence, read sequence, result sequence.
In other embodiments, as shown in fig. 3, the instruction parsing module specifically includes an instruction arbiter, an instruction buffer queue, an instruction parsing unit, and an accelerator arbiter.
The number of the instruction buffer queues and the number of the instruction analysis units are respectively a plurality, the instruction buffer queues correspond to the instruction analysis units one by one, and in the embodiment, the number of the instruction buffer queues and the number of the instruction analysis units are respectively 5. The instruction buffer queue is respectively connected with the instruction arbitrator and the instruction analyzing unit, the instruction analyzing unit is connected with the accelerator arbitrator, and the accelerator arbitrator is connected with the accelerator array module.
The instruction arbitrator is used for receiving the instruction information from the general computing module and selecting an instruction cache queue for the instruction information. In this embodiment, the selected indicator is the congestion level of the instruction cache queue, i.e., the instruction cache queue with the lowest congestion level is selected. The instruction buffer queue is used for buffering the instruction information and distributing the instruction information to the corresponding instruction analysis unit. In this embodiment, if the current instruction information is analyzed completely, the next instruction information in the instruction buffer queue may be analyzed, and if the current instruction information is not analyzed completely, the next instruction information is waited for, and at this time, the next instruction information continues to be in the instruction buffer queue.
The number of the instruction buffer queues and the instruction analysis units in the instruction analysis module is several, so that the instruction information can be processed in parallel, and the high-parallelism characteristic is realized. The instruction analysis module is arranged as an interactive interface between the general computation module and the accelerator array module on the whole, so that the general computation module can send instruction information to the accelerator array module in a non-blocking manner, and the purpose of supporting batch instruction operation is achieved.
The instruction analysis unit is a core part of the instruction analysis module design. The instruction parsing unit is configured to parse the instruction information, specifically, the instruction parsing unit is configured to process a first preset segment of the instruction information sent by the general computation module to the accelerator array module, where the first preset segment includes rs1 and rs2 in this embodiment. And the second preset segment is also used for processing the instruction information sent by the accelerator array module to the general computation module, in this embodiment, the second preset segment includes rd, and in other embodiments, the second preset segment may also include data. And also for processing cache information sent by the accelerator array module to the general purpose computing module, which in this embodiment includes input sequences for the dynamic programming algorithm (two in this embodiment, i.e., the reference sequence and the compare sequence), and parameters for the dynamic programming computation.
The accelerator arbiter is an interactive port of the instruction analysis unit and the accelerator array module and is used for asynchronously distributing the analysis result of the instruction analysis unit to the accelerator array module, and the accelerator array module is used for performing dynamic planning calculation according to the analysis result; the accelerator arbiter is also used for asynchronously reading the data completed by the accelerator array module.
The instruction information includes one or more of data instructions, parameter instructions, and calculation instructions, all of which are included in this embodiment. In other embodiments, auxiliary instructions may also be included. The data instruction is used for configuring the accelerator array module to calculate required data addresses, single data size and total data length, and the calling times of the data instruction depend on which types of data are required by calculation. The parameter instructions are used for configuring parameters of the dynamic programming calculation of the accelerator array module, and are usually called once. The calculation instruction is used for starting the accelerator array module to execute dynamic programming calculation on the data after the data preparation is finished, and writing the calculation result back to the data address configured by the data instruction. The auxiliary instructions are used for detecting the behaviors of the accelerator array module, such as judging whether the current accelerator completes the calculation or not.
As shown in fig. 4, the accelerator array module includes a granularity configuration module, an input buffer sequence, and m computing units. The m computing units form a rectangular array, in this embodiment, the number of computing units is 9, and a rectangular array of 3 × 3, specifically a systolic array, is formed. The input buffer sequence is used for storing the read sequence, and the input buffer sequence is used for storing the reference sequence.
As shown in fig. 5, the computing unit includes a data selector, a character comparison module, a result temporary storage queue, a backtracking logic module, and a data buffer queue.
The number of the character comparison modules is a plurality, the number of the result temporary storage queues is consistent with that of the character comparison modules, and the result temporary storage queues correspond to the character comparison modules one by one.
The data selector is used for receiving the input reference sequence and the comparison sequence which is calculated in the previous round in the data buffer queue and selecting data; in this embodiment, the data selection is determined according to the requirement of the character comparison module, in other words, the data selection is performed according to whether the calculation data required by the character comparison module is the alignment sequence of the previous round or the input reference sequence.
The character comparison module is a core calculation module and is used for comparing the read sequence with a reference sequence or a comparison sequence input by the data selector and storing a comparison result in a result temporary storage queue. For example, the four character comparison module in this embodiment would perform a comparison calculation of four data in the read sequence. That is, each iteration compares four data, several comparisons until the read sequence is completed.
After the comparison calculation between the read sequence and the reference sequence and between the read sequence and the comparison sequence is completed, the backtracking logic module is used for extracting the comparison result in the result temporary storage queue to perform reverse backtracking of dynamic planning to obtain the comparison sequence, and the comparison sequence is stored in the data cache queue for the next calculation.
The granularity configuration module is used for integrating n computing units into an accelerator to realize configuration of the granularity of the array, wherein n is smaller than or equal to m. Because the length of the read sequence is different from one sequencing technology to another and from one company to another, a single calculation unit cannot calculate the read sequences with different lengths, and the calculation of longer read sequences can be completed by integrating a plurality of small calculation units into an accelerator. The granularity configuration module integrates the n calculation units into the accelerator, so that the adjustment of the granularity of the n number of the n of the granularity of the reading of the granularity of the reading of the granularity.
The scheme gets rid of bandwidth bottleneck in the interaction process caused by the existing interface and bus, so that the universal computing module and the accelerator array module directly interact through instruction information, and interaction loss is reduced. Moreover, batch instruction operation can be efficiently carried out through the custom instruction, only the corresponding custom instruction needs to be independently adjusted for different types of tasks, the instruction set does not need to be integrally adjusted, and the complexity of instruction set design is simplified. The data instruction, the parameter instruction and the calculation instruction are distinguished clearly, pertinence is strong, and accurate scheduling of the accelerator array module is facilitated. Meanwhile, a special customized operation function module is adopted, and a dynamic algorithm is executed through a special hardware structure, so that the execution efficiency of the steps of BWA-MEM and Indel reading in gene sequencing is improved.
The in-memory computing unit is used for calling and executing a preset in-memory sequencing operator, and the in-memory sequencing operator is a sequencing operator with high concurrency characteristics and comprises an FM index operator or a Hash index operator. Specifically, the memory computing unit comprises a main memory module and a logic computing module, the main memory module is used for caching gene data in the gene sequencing process, the main memory module utilizes a volatile memory device and is used for temporarily storing intermediate data used in the gene sequencing process, and the intermediate data is gene data related to the gene sequencing process. In this embodiment, the main memory module adopts an SDRAM memory array to form the main memory of the whole computer system. The central processing unit and the storage computing unit access data in the main memory module by accessing the DIMM, and the central control module in the central processing unit can also issue configuration and management information to the logic computing module through the DIMM.
The logic calculation module is used for calling the gene data in the main memory module, calling a preset in-memory sequencing operator and executing the in-memory sequencing operator to process the called gene data, and particularly, the intermediate data is processed according to an FM index operator or a Hash index operator, so that the data calculation of gene sequencing is completed in the data access process. In this embodiment, the logic calculation module uses a dedicated hardware structure supporting Hash calculation to offload the Hash Index and FM Index in the BWA-MEM step in the gene sequencing process to the logic calculation module, and the logic control module is configured to call the required data from the main memory module to execute the Hash Index or FM Index when receiving the management information.
In the scheme, the Hash Index and FM Index logics are realized by adopting special hardware, compared with the traditional software, the processing speed is higher, and the data processing performance in the gene sequencing process is improved. And meanwhile, near data processing is adopted, so that the data moving overhead is reduced, and the execution efficiency of the BWA-MEM step is further improved.
The storage computing unit is used for calling and executing a preset storage sequencing operator, and the storage sequencing operator comprises a sequencing operator, a compression/decompression operator and a redundancy removing operator. Specifically, the storage computing unit comprises a storage module and a basic processing module. The storage module is used for storing gene data in a gene sequencing process; and also for storing gene data for completing gene sequencing. The storage module utilizes a nonvolatile storage device and is used for permanently storing intermediate data or gene data for finishing gene sequencing, wherein the intermediate data is the gene data used in a gene sequencing process. In the embodiment, the storage module adopts an SSD.
The basic processing module is used for calling the gene data in the storage module, calling a preset storage sequencing operator and executing the storage sequencing operator to process the called gene data. And processing the intermediate data according to the sequencing operator, the compression/decompression operator or the redundancy removing operator, thereby realizing the data processing of gene sequencing in the data access process. Specifically, the basic processing module adopts a programmable hardware logic unit, and unloads the Sort step in the gene sequencing process and the storage compression and decompression processes required among all the gene sequencing steps to the basic processing module, and in this embodiment, the basic processing module adopts an FPGA.
The storage module and the basic processing module are arranged in the storage computing unit together, gene data required by the basic processing module for processing are stored in the storage module, the physical distance between the storage module and the basic processing module is shortened, the data moving cost is reduced, and the data processing performance of the storage computing unit is improved. Meanwhile, certain steps in the gene sequencing process are unloaded to the programmable basic processing module, and compared with the traditional CPU processing, the time overhead of switching among the steps in the gene sequencing process is reduced. For example, the compression step is unloaded to the basic processing module, so that the gene data is compressed in the stored process, high concurrent processing of the compression process is realized, and compared with the gene sequencing process executed in the traditional CPU, the time overhead caused by data compression and decompression during switching among steps is reduced.
Taking compression and decompression as an example, after each gene sequencing step is executed, the central processing unit controls the calculated gene data to be sent to the storage calculation module, the storage calculation module sends the gene data to the basic processing module while receiving the gene data, and the basic processing unit executes a compression operator to compress the gene data and then stores the compressed gene data. When each gene sequencing step starts to be executed, the central processing unit initiates a data reading request, reads gene data in the storage module to a main memory module of the memory computing unit, and the memory computing unit executes a decompression operator on the data to be read through the basic processing module and then sends the data to the memory computing unit.
In the scheme, the compression and decompression steps are unloaded to the programmable hardware logic unit, the hardware unloading compression process is used, high concurrency processing of the compression process is realized, data is compressed in the stored process, and time overhead caused by data compression and decompression in the switching process of the traditional gene sequencing process is reduced.
The present embodiment also provides a gene sequencing method, using the gene sequencing system, connecting the central processing unit, the memory computing unit and the storage computing unit through the root complex, and processing gene data through the central processing unit, the memory computing unit and the storage computing unit. The method comprises the following steps:
a communication linkage step: and establishing a communication link between the root complex and the central processing unit, the memory computing unit and the storage computing unit, wherein the communication comprises data transmission of gene data and interaction of management messages. The method specifically comprises the following steps:
the central processing unit is connected with the root complex through the FSB, sends control and management messages to the memory computing unit and the storage computing unit, receives completion events, sends and receives gene data, and achieves management and communication of the memory computing unit and the storage computing unit.
The root complex is connected with the memory computing unit through the DIMM, and the root complex is connected with the memory computing unit through the DIMM and interacts data and management messages with the memory computing unit. The root complex is connected with the storage computing unit through PCIe, and DMA data and control messages can be mutually sent and received between the storage computing unit and the storage computing unit.
The central processing unit, the memory computing unit and the storage computing unit are managed in a centralized way through the root complex, and mutual communication among the units is realized.
A central processing step: and calling and executing a preset central sequencing operator through a central processing unit, wherein the central sequencing operator comprises a dynamic programming sequencing operator and a hidden Markov model sequencing operator. The method specifically comprises the following steps:
in the gene sequencing process, gene data are processed according to a dynamic programming sequencing operator and a hidden Markov model sequencing operator. In other embodiments, the central processing unit manages the gene sequencing process, and in particular manages the gene sequencing process and the programming interface for gene sequencing.
Executing the dynamic programming operator includes the following:
splitting an instruction set of a dynamic programming algorithm, and distributing specific instruction information in the instruction set; the instruction information comprises one or more of a data instruction, a parameter instruction and a calculation instruction; analyzing the instruction information, and distributing the analyzed instruction information; and controlling an accelerator array module to execute dynamic programming calculation according to the instruction information.
An in-memory calculation step: and calling and executing a preset in-memory sequencing operator through the in-memory computing unit, wherein the in-memory sequencing operator is a sequencing operator with high concurrency characteristics and comprises an FM index operator or a Hash index operator. The method specifically comprises the following steps:
the memory computing unit comprises a main memory module and a logic computing module.
The main memory module stores the gene data in the gene sequencing process, and temporarily stores intermediate data used in the gene sequencing process by using a volatile memory device, wherein the intermediate data is the gene data related to the gene sequencing process. In the embodiment, the main memory module adopts SDRAM memory array.
The method comprises the steps of calling gene data stored in a main memory module, calling a preset in-memory sequencing operator by a logic calculation module, and executing the in-memory sequencing operator to process the called gene data, specifically processing intermediate data according to an FM index operator or a Hash index operator, thereby realizing the data calculation of gene sequencing in the data access process. In this embodiment, the logic calculation module uses a dedicated hardware structure supporting Hash calculation to offload the Hash Index and FM Index in the BWA-MEM step in the gene sequencing process to the logic calculation module, and the logic control module is configured to call the required data from the main memory module to execute the Hash Index or FM Index when receiving the management information.
And a storage calculation step: and calling and executing a preset storage sequencing operator through the storage computing unit, wherein the storage sequencing operator comprises a sequencing operator, a compression/decompression operator and a redundancy removing operator. The method specifically comprises the following steps:
the storage computing unit comprises a storage module and a basic processing module.
And storing the gene data in the gene sequencing process and the gene data for finishing the gene sequencing through a storage module. The storage module utilizes a nonvolatile storage device to permanently store intermediate data or gene data for completing gene sequencing, wherein the intermediate data is the gene data used in the gene sequencing process. In the embodiment, the storage module adopts an SSD.
And calling the gene data stored in the storage module, calling a preset storage sequencing operator by the basic processing module, and executing the storage sequencing operator to process the called gene data. The intermediate data is processed according to a sequencing operator, a compression/decompression operator or a redundancy removing operator, so that the data processing of gene sequencing is completed in the data access process.
The central processing step, the memory computing step and the storage computing step can be simultaneously operated in parallel, for example, the gene sequencing process comprises a step A, a step B and a step C, the three steps are respectively operated in the central processing unit, the memory computing unit and the storage computing unit, and when the memory computing unit processes the step B in the current gene sequencing process, the central processing unit processes the step A in another gene sequencing process in parallel. Through the parallel processing of a plurality of units, the parallel efficiency among different processing steps of gene sequencing is effectively improved.
Taking compression and decompression as an example, after each gene sequencing step is executed, the central processing unit controls the calculated gene data to be sent to the storage calculation module, the storage calculation module sends the gene data to the basic processing module while receiving the gene data, and the basic processing unit executes a compression operator to compress the gene data and then stores the compressed gene data. When each gene sequencing step starts to be executed, the central processing unit initiates a data reading request, reads gene data in the storage module to a main memory module of the memory computing unit, and the memory computing unit executes a decompression operator on the data to be read through the basic processing module and then sends the data to the memory computing unit.
By adopting the scheme, the compression and decompression steps are unloaded to the programmable hardware logic unit, the hardware unloading compression process is used, the high concurrency processing of the compression process is realized, the data is compressed in the stored process, and the time overhead caused by data compression and decompression in the switching process of the traditional gene sequencing process is reduced.
Example two
The difference between the present embodiment and the first embodiment is:
a gene sequencing system is characterized in that an in-memory computing unit and a storage computing unit both comprise a flow control module, wherein the flow control module is used for receiving flow management information, arbitrating the flow management information to obtain a management command and distributing the management command to a preset management message queue; the management device is also used for receiving the data information and calling the management command in the corresponding management message queue according to the data information; executing the management command; the memory computing unit and the storage computing unit are also used for calling and executing the corresponding memory sequencing operator or storage sequencing operator when executing the management command. Specifically, the method comprises the following steps:
as shown in fig. 6, the process control module includes a multi-queue submodule, a management message arbitration submodule, and a circulation submodule, where the multi-queue submodule is configured to store a plurality of management message queues and context information of each management message queue, where the context information refers to information of the management message queue, and includes a queue number and a data depth, that is, the management message queue includes a queue number. The setting of a plurality of management message queues supports message interfaces of a plurality of queues, and realizes concurrent processing of a plurality of gene sequencing processes, thereby reducing the time overhead in the gene sequencing processes.
The management message arbitration submodule is used for receiving process management information, and the process management information is transmitted from the outside, such as process management software. The management message arbitration sub-module is also used for arbitrating the process management information to obtain a management command, analyzing the process management information to obtain a message queue number, and writing the management command into the corresponding management message queue in sequence according to the message queue number and the queue number. The management message queue contains a plurality of management commands, each management command comprises a message operation code and an operation code parameter, the message operation code marks the operation to be executed by the data, and the operation comprises executing a gene sequencing operator, reading or writing data from or to the local, and reading or writing data from or to the remote, namely the management commands comprise executing the gene sequencing operator, reading data and writing data, the reading data comprises reading data from the local and reading data from the remote, and the writing data comprises writing data to the local and writing data to the remote. The opcode parameter is to provide sideband information required for operation, such as executing a management command to read data from a local area, and the address and size of the local area data are required to be known.
The data circulation submodule is used for receiving data information, and the data information is transmitted from the outside or prestored by the system and obtained by adopting a calling mode. The data circulation sub-module is also used for analyzing the data information to obtain a data queue number, screening a corresponding management message queue according to the data queue number and the queue number, and calling a management command in the screened management message queue. The multi-queue submodule is also used for deleting the corresponding management command in the management message queue when the management command in the management message queue is called. And after the management command is called, deleting the corresponding management command, so that the management command executed in the next gene sequencing process is positioned at the first position of the management message queue, and realizing the quick calling of the management command when the next data information comes.
The flow control module is also used for judging whether the management command is write-in data after executing the management command, waiting for next data information when the management command is the write-in data, and otherwise, calling the management command in the management message queue according to the data information after executing the management command and executing the management command.
The memory computing unit and the storage computing unit are also used for calling and executing the corresponding memory sequencing operator or storage sequencing operator when executing the management command. For example, when the management command executed by the flow control module in the in-memory computing unit is to execute a seed operation in a gene sequencing process, the logic computing module is configured to call the gene data in the main memory module, call an in-memory sequencing operator required for executing the seed operation according to the management command, and execute the in-memory sequencing operator to process the called gene data, thereby obtaining processed gene data.
By adopting the scheme, the process control module is arranged in the memory computing unit and the storage computing unit, the gene sequencing process in each unit is managed through the process control module, and the centralized management of the gene sequencing process is realized through the programmable management of the process management information. The processing and transmission of the data information are unloaded to the memory computing unit and the storage computing unit for processing, so that the delay of data information control is reduced, the time overhead of the system in the control of the gene sequencing process is effectively reduced, and the efficient control of the gene sequencing process is realized.
In addition, the present embodiment also provides a gene sequencing method, using the gene sequencing system, further including the following steps:
command management step: and receiving the flow management information, arbitrating the flow management information to obtain a management command, and distributing the management command to a preset management message queue. The command management step specifically comprises the following steps:
process management information is received, which is transmitted from an external source, such as process management software.
And arbitrating the flow management information to obtain a management command, analyzing the flow management information to obtain a message queue number, and writing the management command into the corresponding management message queue in sequence according to the message queue number and the queue number.
The preset management message queue is multiple, the management message queue after the management command is written comprises multiple management commands, each management command comprises a message operation code and an operation code parameter, the message operation code marks the operation to be executed by the data, and comprises the steps of executing a gene sequencing operator, reading or writing data from or to the local, and reading or writing data from or to the remote, namely the management command comprises the steps of executing the gene sequencing operator, reading the data and writing the data, the reading the data comprises reading the data from the local and reading the data from the remote, and the writing the data comprises writing the data to the local and writing the data to the remote. The opcode parameter is to provide sideband information required for operation, such as executing a management command to read data from a local area, and the address and size of the local area data are required to be known.
A command execution step: and receiving the data information, calling a management command in the management message queue according to the data information, and calling an in-memory calculation step or a storage calculation step to execute the management command. Context information of each management message queue is also preset, and the context information refers to information of the management message queue and comprises a queue number and data depth, namely the management message queue comprises the queue number. The command execution step specifically comprises the following steps:
and receiving data information, wherein the data information is transmitted from the outside or prestored by the system and is obtained by adopting a calling mode.
Analyzing the data information to obtain a data queue number, screening a corresponding management message queue according to the data queue number and the queue number, calling a management command in the screened management message queue, and deleting the corresponding management command in the management message queue.
And calling the in-memory calculation step or the storage calculation step to execute management commands, wherein each management command comprises a message operation code and an operation code parameter, the message operation code marks the operation to be executed by the data, and the operation code comprises executing a gene sequencing operator, reading or writing data from or into the local, and reading or writing data from or into the remote, namely the management command comprises executing the gene sequencing operator, reading data and writing data, the reading data comprises reading data from the local and reading data from the remote, and the writing data comprises writing data to the local and writing data to the remote. The opcode parameter is to provide sideband information required for operation, such as executing a management command to read data from a local area, and the address and size of the local area data are required to be known.
Executing a judging step: and after the management command is executed, judging whether the management command is write-in data, waiting for next data information when the management command is the write-in data, and otherwise, calling a command execution step according to the data information after the management command is executed.
By adopting the scheme, programmable gene sequencing flow control is provided through flow management information, centralized management of the gene sequencing flow is realized, and time overhead of the system in the gene sequencing flow control is effectively reduced.
The foregoing is merely an example of the present invention, and common general knowledge in the field of known specific structures and characteristics is not described herein in any greater extent than that known in the art at the filing date or prior to the priority date of the application, so that those skilled in the art can now appreciate that all of the above-described techniques in this field and have the ability to apply routine experimentation before this date can be combined with one or more of the present teachings to complete and implement the present invention, and that certain typical known structures or known methods do not pose any impediments to the implementation of the present invention by those skilled in the art. It should be noted that, for those skilled in the art, without departing from the structure of the present invention, several changes and modifications can be made, which should also be regarded as the protection scope of the present invention, and these will not affect the effect of the implementation of the present invention and the practicability of the patent. The scope of the claims of the present application shall be determined by the contents of the claims, and the description of the embodiments and the like in the specification shall be used to explain the contents of the claims.
Claims (10)
1. A gene sequencing system, comprising: the system comprises a central processing unit, an in-memory computing unit and a storage computing unit, wherein the central processing unit, the in-memory computing unit and the storage computing unit can simultaneously run in parallel, the central processing unit is used for calling and executing a central sequencing operator, the in-memory computing unit is used for calling and executing an in-memory sequencing operator, and the storage computing unit is used for calling and executing a storage sequencing operator.
2. A gene sequencing system according to claim 1, wherein: an in-memory computing unit comprising:
the main memory module is used for caching gene data in a gene sequencing process;
and the logic calculation module is used for calling the gene data in the main memory module, calling a preset in-memory sequencing operator and executing the in-memory sequencing operator to process the called gene data.
3. A gene sequencing system according to claim 1, wherein: the central processing unit comprises a central control module and a customized operation function module which can run in parallel at the same time, the customized operation function module is used for calling and executing a central sequencing operator, and the central processing module is used for calling and executing other sequencing operators except the central sequencing operator, the in-memory sequencing operator and the storage sequencing operator.
4. A gene sequencing system according to claim 1, wherein: the memory computing unit and the memory computing unit both comprise a flow control module;
the flow control module is used for receiving the flow management information, arbitrating the flow management information to obtain a management command, and distributing the management command to a preset management message queue; the management device is also used for receiving the data information, calling the management command in the corresponding management message queue according to the data information and executing the management command;
the memory computing unit and the storage computing unit are also used for calling and executing the corresponding memory sequencing operator or storage sequencing operator when executing the management command.
5. A gene sequencing system according to claim 4, wherein: the management message queue comprises a queue number, the process control module comprises a management message arbitration submodule, and the management message arbitration submodule is used for analyzing the process management information to obtain the message queue number and writing the management commands into the corresponding management message queue in sequence according to the message queue number and the queue number.
6. A gene sequencing system according to claim 4, wherein: the management message queue comprises a queue number, the process control module comprises a data flow submodule and a multi-queue submodule, the data flow submodule is used for analyzing data information to obtain the data queue number, the corresponding management message queue is screened according to the data queue number and the queue number, a management command in the screened management message queue is called, and the multi-queue submodule is used for deleting the corresponding management command in the management message queue when the management command in the management message queue is called.
7. A method of gene sequencing, comprising: the method comprises the following steps:
a central processing step: calling and executing a preset central sequencing operator through a central processing unit;
an in-memory calculation step: calling and executing a preset in-memory sequencing operator through an in-memory computing unit;
and a storage calculation step: calling and executing a preset storage sequencing operator through a storage computing unit;
the central processing step, the in-memory computing step and the storage computing step may be run concurrently and in parallel.
8. A gene sequencing system according to claim 7, wherein: the memory computing step and the storage computing step comprise the following steps:
storing gene data in a gene sequencing process;
and calling the stored gene data, calling a preset in-memory sequencing operator or a storage sequencing operator, and executing the in-memory sequencing operator or the storage sequencing operator to process the called gene data.
9. A method of gene sequencing according to claim 7, wherein: the method also comprises the following steps:
command management step: receiving flow management information, arbitrating the flow management information to obtain a management command, and distributing the management command to a preset management message queue;
a command execution step: and receiving the data information, calling a management command in the management message queue according to the data information, and calling an in-memory calculation step or a storage calculation step to execute the management command.
10. A method of gene sequencing according to claim 9, wherein: the management message queue comprises a queue number and distributes management commands to a preset management message queue, and the management command queue comprises the following contents:
analyzing the flow management information to obtain a message queue number, and writing the management commands into the corresponding management message queues in sequence according to the message queue number and the queue number;
invoking a management command in the management message queue according to the data information, wherein the management command comprises the following contents:
analyzing the data information to obtain a data queue number, screening a corresponding management message queue according to the data queue number and the queue number, calling a management command in the screened management message queue, and deleting the corresponding management command in the management message queue.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110632394.2A CN113241120A (en) | 2021-06-07 | 2021-06-07 | Gene sequencing system and sequencing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110632394.2A CN113241120A (en) | 2021-06-07 | 2021-06-07 | Gene sequencing system and sequencing method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113241120A true CN113241120A (en) | 2021-08-10 |
Family
ID=77137040
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110632394.2A Pending CN113241120A (en) | 2021-06-07 | 2021-06-07 | Gene sequencing system and sequencing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113241120A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114276896A (en) * | 2021-12-22 | 2022-04-05 | 成都瀚辰光翼科技有限责任公司 | Automatic nucleic acid extraction system, method and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107370667A (en) * | 2017-07-31 | 2017-11-21 | 北京北信源软件股份有限公司 | Multi-threading parallel process method and apparatus, computer-readable recording medium and storage control |
| CN109785905A (en) * | 2018-12-18 | 2019-05-21 | 中国科学院计算技术研究所 | A kind of accelerator towards gene alignment algorithm |
| CN111653317A (en) * | 2019-05-24 | 2020-09-11 | 北京哲源科技有限责任公司 | Gene comparison accelerating device, method and system |
| WO2021011920A1 (en) * | 2019-07-18 | 2021-01-21 | Micron Technology, Inc. | Arbiter circuit for commands from multiple physical functions in a memory sub-system |
-
2021
- 2021-06-07 CN CN202110632394.2A patent/CN113241120A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107370667A (en) * | 2017-07-31 | 2017-11-21 | 北京北信源软件股份有限公司 | Multi-threading parallel process method and apparatus, computer-readable recording medium and storage control |
| CN109785905A (en) * | 2018-12-18 | 2019-05-21 | 中国科学院计算技术研究所 | A kind of accelerator towards gene alignment algorithm |
| CN111653317A (en) * | 2019-05-24 | 2020-09-11 | 北京哲源科技有限责任公司 | Gene comparison accelerating device, method and system |
| WO2021011920A1 (en) * | 2019-07-18 | 2021-01-21 | Micron Technology, Inc. | Arbiter circuit for commands from multiple physical functions in a memory sub-system |
Non-Patent Citations (2)
| Title |
|---|
| ZEKUN YIN 等: ""Computing Platforms for Big Biological Data Analytics: Perspectives and Challenges"", 《COMPUTATIONAL AND STRUCTURAL BIOTECHNOLOGY JOURNAL》, vol. 15, pages 403 - 411 * |
| 王元戎 等: ""Gene Panel流程的并行设计与优化研究"", 《计算机学报》, vol. 42, no. 11, pages 2429 - 2446 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114276896A (en) * | 2021-12-22 | 2022-04-05 | 成都瀚辰光翼科技有限责任公司 | Automatic nucleic acid extraction system, method and storage medium |
| CN114276896B (en) * | 2021-12-22 | 2023-12-29 | 成都瀚辰光翼科技有限责任公司 | Automatic nucleic acid extraction method and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA2313462C (en) | Multiprocessor computer architecture incorporating a plurality of memory algorithm processors in the memory subsystem | |
| US20220382460A1 (en) | Distributed storage system and data processing method | |
| JP2002149424A (en) | A plurality of logical interfaces to shared coprocessor resource | |
| EP2131278A1 (en) | Scheduling of multiple tasks in a system including multiple computing elements | |
| JPH0844651A (en) | Input/output control system with plurality of channel passeswith reference to input/output device | |
| JPS6258341A (en) | Input and output interruption processing system | |
| CN102934102A (en) | Multiprocessor system, execution control method and execution control program | |
| CN114610472B (en) | Multi-process management method in heterogeneous computing and computing equipment | |
| CN113241120A (en) | Gene sequencing system and sequencing method | |
| JP2000293436A (en) | Support for a plurality of unsolved request to a plurality of targets of pipeline memory system | |
| US9286129B2 (en) | Termination of requests in a distributed coprocessor system | |
| CN112789593A (en) | Multithreading-based instruction processing method and device | |
| CN113268270A (en) | Acceleration method, system and device for paired hidden Markov models | |
| JPH06131181A (en) | Hierarchical parallel processing type central processing unit | |
| CN113268269B (en) | Acceleration method, system and device for dynamic programming algorithm | |
| US9952991B1 (en) | Systematic method on queuing of descriptors for multiple flash intelligent DMA engine operation | |
| US20230393782A1 (en) | Io request pipeline processing device, method and system, and storage medium | |
| TW202242638A (en) | Instruction dispatch for superscalar processors | |
| CN113139798B (en) | Gene sequencing flow management control method and system | |
| US20040128476A1 (en) | Scheme to simplify instruction buffer logic supporting multiple strands | |
| CN109388596A (en) | A kind of data manipulation method and device | |
| JPH02245864A (en) | Multiprocessor system | |
| CN116755868B (en) | Task processing system and method | |
| CN1097783C (en) | System and method for processing of data and a communications system with such a system | |
| US12001681B2 (en) | Distributed storage system and data processing method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |