Data stream identification method and device

Abstract

The invention provides a data stream recognition method and a device. The method comprises the steps of: receiving a host request, wherein the host request comprises at least one data stream; allocating a sliding window to each data stream in the host request, wherein the sliding window has a start logic address and an end logic address; updating the sliding window allocated to the data stream according to a data block requested to be accessed in the data stream; and within first preset time, if the start logic address and the end logic address of the sliding window are updated and the size of the sliding window is not exceed a preset maximum value, recognizing the data stream as a progressive stream. The device comprises a receiving module, an allocation module, an updating module and a recognition module. By adopting the scheme provided by the invention, the out-of-order problem can be effectively solved and the recognition of multiple progressive streams can be effectively realized.

Description

technical field [0001] Embodiments of the present invention relate to the field of data storage, and in particular, to a data flow identification method and device. Background technique [0002] With the wide application of streaming media, the read performance requirements of the storage system are getting higher and higher. To improve the read performance of the storage system, it is necessary to pre-judge the host read request, that is, before the host read request arrives, the data to be read by the host has been pre-read from the disk into the cache (CACHE), so that Host read requests can be directly hit in the cache, thereby shortening the latency of host read requests and improving the read performance of the storage system. [0003] To improve the hit rate of host read requests in the cache and shorten the delay of read requests, the premise is that the system must support the prefetch function, and the key to prefetch is to identify the characteristics of input and...

AI Technical Summary

Problems solved by technology

[0006] The host sends requests in order, but due to the possibility of out-of-order during transmission, the requests are not necessarily in order when they arrive at the storage system. This method does not consider the out-of-order of the sequence flow, nor does it consider the multi-way order The coexistence of streams makes the current data flow identification methods unable to identify out-of-order sequential streams, and cannot deal with the coexistence of most sequential streams, which directly leads to low efficiency and poor accuracy of sequential stream identification.

Images