IO current limiting method and device based on heterogeneous storage and medium

Abstract

The embodiment of the invention discloses an IO current limiting method and device based on heterogeneous storage and a medium. The method comprises the steps: setting the maximum resource distribution amount of a volume, an I / O queue depth threshold and an I / O queue horizontal line according to the service indexes of storage equipment; and calculating the exclusive I / O resource quantity and the shared I / O resource quantity corresponding to the channels according to the maximum resource allocation quantity of the volume and the number of the channels of the storage device. Dynamically adjusting I / O resource quantities corresponding to the volumes and the channels according to the number of the volumes and the number of the channels; According to the volume activity I / O quantity of the volume, the I / O queue depth threshold and the I / O queue horizontal line, allocating additional I / O resource quantity to the volume and the channel. Counting the channel activity I / O number of the target channel; if the number of the active I / Os of the channel does not meet the resource release rule, the new I / O task is put into the waiting queue, so that the stability of heterogeneous scene services can be greatly improved, the IO flow limiting effect is improved, and the QoS problem is effectively reduced.

Description

technical field [0001] The present invention relates to the technical field of storage systems, in particular to a heterogeneous storage-based IO current limiting method, device and computer-readable storage medium. Background technique [0002] As an advanced feature of a storage system, heterogeneous storage can take over volumes mapped from other storage systems to the current storage system to integrate storage resources. Due to the large number of existing mainstream storage systems, including the series and models of storage systems within the same manufacturer, it is difficult to manage back-end storage functions. [0003] In actual application, it is found that due to the different performance and stability of the back-end storage devices, the front-end main storage system is prone to problems related to Quality of Service (QoS) when processing IO, including: storage system resource allocation Insufficient, resulting in insufficient utilization of the actual perform...

AI Technical Summary

Problems solved by technology

Due to the large number of existing mainstream storage systems, including the series and models of storage systems within the same manufacturer, this has caused certain difficulties in managing back-end storage functions

[0003] In actual application, it is found that due to the different performance and stability of the back-end storage devices, the front-end main storage system is prone to problems related to Quality of Service (QoS) when processing IO, including: storage system resource allocation Insufficient, resulting in insufficient utilization of the actual performance of the back-end storage; insufficient IO current limiting control work of the storage system leads to IO congestion, heterogeneous devices or volumes are offline, intermittent, or IO timeouts due to excessive queuing; when there are multiple back-end storage devices or multiple volumes, the overall performance and stability are affected due to uneven or too even resource allocation

Images