Browse Prior Art Database

Method And System For Managing One Or More Solid State Drives (SSD) Devices

IP.com Disclosure Number: IPCOM000201809D
Publication Date: 2010-Nov-23
Document File: 3 page(s) / 56K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a method and system for managing one or more Solid State Drives (SSD) devices. The method monitors each SSD device periodically to determine an error-count corresponding to each SSD device. Based on the error-count for a SSD device, the method directs subsequent write commands to a temporary SSD device. Additionally, a new SSD device is identified for replacing the write-failed SSD device. Accordingly, file contents from the write-failed SSD device are merged to the new SSD device.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 3

Method And System For Managing One Or More Solid State Drives (SSD) Devices

A method and system is disclosed for managing one or more Solid State Drives (SSDs) devices. The method monitors each SSD devices periodically to determine an error-count corresponding to each SSD device. Based on the error-count for a SSD device, the method directs subsequent write commands to a temporary SSD device. Additionally, a new SSD device is identified for replacing the write-failed SSD device. Accordingly, file contents from the write-failed SSD device are merged to the new SSD device.

Fig. 1 illustrates the method steps which are executed for monitoring each SSD device.

(This page contains 00 pictures or other non-text object)

Figure 1

In accordance with the method and system disclosed herein, a user set a polling time for each SSD device to a default value. In an embodiment, the default value of 60 seconds may be used. The polling time is set by taking into account several storage area network unique factors such number of SSD devices and reliability a user requires versus performance. Further, the user may also set a threshold for write-errors for each SSD device. Based on the threshold for write-errors, a SSD device may not be used once an error-count for the SSD device crosses the threshold.

The method keeps monitoring an amount of data going to each SSD device and accordingly adjusts a timer which controls when SCSI log sense command is issued. In

1


Page 02 of 3

an embodiment, SSD devices which are lightly-used are issued the SCSI log sense command less frequently as compared to heavily-used SSD devices.

Moving on, the method issues a SCSI log-sense command with page-code set to error-counts for each SSD device which is being monitored. Accordingly, error-counts that are returned by each SSD device are checked to determine if the error-counts for a SSD device has crossed the threshold specified by the user. If the threshold is crossed, it is established that the SSD device is write degraded. Thereafter, for each write-failed SSD device, a file mergi...