Accelerating device for gene comparison algorithm

Abstract

The invention provides an accelerating device for a gene comparison algorithm. The accelerating device comprises a host terminal and a plurality of accelerating modules. Each accelerating module comprises a plurality of storage layers, a vertical slice management unit and a switching network. The host terminal is used for controlling distribution of gene sequencing sequences to the accelerating modules and receiving of gene comparison results. The storage layers of the accelerating modules are used for storing gene reference sequences. The vertical slice management units of the accelerating modules are used for managing slices formed by vertically dividing the plurality of storage layers and executing a gene comparison algorithm to obtain a gene comparison result. The switching networks ofthe accelerating modules are used for controlling data exchange inside the accelerating modules and data exchange between the accelerating modules and the outside. The accelerating device disclosed by the invention can improve the processing speed of a gene comparison algorithm by utilizing a customized structure.

Description

technical field [0001] The invention relates to the technical field of biological gene data processing, in particular to an acceleration device oriented to a gene comparison algorithm. Background technique [0002] At present, the development of biotechnology is triggering an industrial revolution in the fields of agriculture, medicine, energy, and materials, effectively alleviating the problems of food, health, and resources, and has a broad space for development. The importance of bioinformatics as a discipline in the biological field is also increasingly prominent. It is one of the major frontier fields of life sciences and natural sciences. Its research focuses on genomics and proteomics, specifically from nucleic acid and protein sequences. As a starting point, the biological information of the structure-function expressed in the sequence is analyzed. In sequence (read) analysis, gene alignment is the most basic and most commonly used method. Through gene alignment, th...

AI Technical Summary

Problems solved by technology

[0003] In the prior art, the gene comparison algorithm usually adopts a general-purpose processing structure (such as a common server, etc.), and there are two main problems in this gene comparison algorithm: 1), the CPU and the memory are usually separated. The memory interface to access data is limited by electrical characteristics, and the memory access bandwidth is limited; in addition, due to the poor locality of the gene comparison operation program, there is a great demand for memory bandwidth. This demand for high bandwidth is different from the current bandwidth. It cannot meet its needs, resulting in poor performance and high power consumption of the gene comparison algorithm

2), general-purpose processors are usually equipped with several functional components such as floating-point arithmetic units, fixed-point arithmetic units, and branch prediction logic, while the gene comparison algorithm has the characteristics of low calculation density and high parallelism, and these components not only occupy processing area of ​​the device silicon, and consumes a large amount of power

Images