Redundant array of inexpensive disks (RAID) system configured to reduce rebuild time and to prevent data sprawl

摘要

A RAID system is provided in which, in the event that a rebuild is to be performed for one of the PDs, a filter driver of the operating system of the computer of the RAID system informs the RAID controller of the RAID system of addresses in the virtual memory that are unused. Unused virtual memory addresses are those which have never been written by the OS as well as those which have been written by the OS and subsequently freed by the OS. The RAID controller translates the unused virtual memory addresses into unused physical addresses. The RAID controller then reconstructs data and parity only for the unused physical addresses in the PD for which the rebuild is being performed. This reduces the amount of data and parity that are rebuilt during a rebuild process and reduces the amount of time that is required to perform the rebuild process. In addition, the RAID system is capable of being configured to prevent or reduce data sprawl.

主权项

1. A Redundant Array of Inexpensive Disks (RAID) system comprising:
a computer comprising:
at least a first processor; anda first local storage system, wherein an operating system (OS) of the computer runs on the first processor and uses the first local storage system, the OS including at least a file system (FS), the FS using a portion of the first local storage system as virtual memory and maintaining a virtual memory table in the first local storage system containing at least entries identifying addresses in the virtual memory that have been written by the FS, wherein the OS causes the virtual memory addresses that have been written by the FS and the corresponding data to be output from the computer; an array of physical disks (PDs), the array having a maximum storage capacity; and a RAID controller interfaced with the computer and with the array of PDs, the RAID controller allocating a first portion of the array for use by the OS that is less than the maximum storage capacity of the array, the RAID controller receiving, over the interface with the computer, at least the virtual memory addresses and the corresponding data, the RAID controller translating the received virtual memory addresses into physical addresses in the PDs, and wherein the RAID controller causes the received data to be written only to physical addresses in the allocated first portion of the array, and wherein if the RAID controller determines that the allocated first portion of the array is close to being filled with data, the RAID controller allocates an additional portion of the array for use by the OS and causes the received data to be written only to physical addresses in the additionally allocated portion of the array.