Server-side, variable drive health determination

• US 9,754,337 B2
• Filed: 04/02/2014
• Issued: 09/05/2017
• Est. Priority Date: 03/29/2012
• Status: Expired due to Fees

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Brooker, Marc J., Parthasarathy, Madhuvanesh, Wei, Danny, Holgers, Tobias L., Li, Yu
• Primary Examiner(s): Vy, Hung T
• Application Number:
  US14/243,752
• Publication Number:
  US 20140214447A1
• Time in Patent Office: 1,252 Days
• Field of Search: 707827, 711170, 711100, 714 2, 709206
• CPC Class Codes: G06F 11/00 Error detection; Error correction; Monitoring error detection, correction or monitoring in information storage based on relative movement between record carrier and transducer G11B20/18; monitoring, i.e. supervising the progress of recording or reproducing G11B27/36; in static stores G11C29/00
• CPC Class Codes: G06F 11/008 Reliability or availability analysis
• CPC Class Codes: G06F 11/0727 in a storage system, e.g. in a DASD or network based storage system drivers for digital recording or reproducing units G06F3/06; circuits for error detection or correction within digital recording or reproducing units G11B20/18; for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS], H04L67/1097
• CPC Class Codes: G06F 11/0793 Remedial or corrective actions recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G06Q 50/22 Social work or social welfare, e.g. community support activities or counselling services

Abstract

The relative health of data storage drives may be determined based, at least in some aspects, on data access information and/or other drive operation information. In some examples, upon receiving the operation information from a computing device, a health level of a drive may be determined. The health level determination may be based at least in part on operating information received from a client entity. Additionally, a storage space allocation instruction or operation may be determined for execution. The allocation instruction or operation determined to be performed may be based at least in part on the determined health level.

Claims

1. One or more computer-readable media collectively storing computer-executable instructions for data storage management that, as a result of being executed by one or more processors, causes the one or more processors to perform operations comprising:

determining, based at least in part on operation information of a storage space received from a device of a computing system, a health level of the storage space, the health level of the storage space being based at least in part on a health scale indicating one or more health levels above a non-operational level, where the operation information indicates one or more attributes of the device;

determining, based at least in part on the health level of the storage space, an instruction for the one or more processors to execute associated with data stored in the storage space;

indicating the health level of the storage space to specify one or more predefined activities defined by the health level for data stored in the storage space; and

performing a remedial operation based at least the one or more predefined activities.

2. The one or more computer-readable media of claim 1, wherein the health scale is based at least in part on at least a healthy level indicating a first failure probability, a suspect level that is less healthy than the healthy level indicating a second failure probability higher than the first failure probability, and an unhealthy level that is less healthy than the suspect level indicating a third failure probability higher than the second failure probability.

3. The one or more computer-readable media of claim 2, wherein the operation information is based at least in part on a write latency associated with the storage space.

4. The one or more computer-readable media of claim 1, wherein the one or more predefined activities further includes storing data stored in the storage space in a second storage space with a health level above a predefined level.

5. The one or more computer-readable media of claim 1, wherein the operation information is based at least in part on a read latency associated with the storage space.

6. The one or more computer-readable media of claim 1, wherein indicating the health level of the storage space comprises at least setting a flag associated with the storage space or indicating the health level in a data structure associated with the storage space to specify the one or more predefined activities that are appropriate for the data stored in the storage space.

7. The one or more computer-readable media of claim 1, wherein the one or more attributes of the device include at least one of:

read latency, write latency, difference between read latency and write latency, packet loss data, spin-up time, seek time, attempts to complete an operation, temperature, software, software version information, or a variance between the one or more attributes of the device.

8. A computer-implemented method comprising:

receiving an indicator associated with a storage device indicating a level of health associated with the storage device to perform operations on data stored in the storage device, the level of health obtained from a plurality of possible health levels included in one or more health levels; and

determining, based at least in part on the indicator, that the level of health associated with the storage device is within an intermediate level of health of the one or more health levels by at least;

maintaining data stored in the storage device;

causing one or more actions to be performed that disable reading or writing to a remaining storage space of the storage device such that no additional data can be read from or written to the storage device; and

associating the storage device within the intermediate level of health in a data store to enable an administrative computer system to utilize the storage device for one or more purposes appropriate, for the intermediate level of health.

9. The computer-implemented method of claim 8, wherein the one or more health levels include at least a first level of health that indicates a low probability of failure, a second level of health that indicates a medium probability of failure that is higher than that of the first level of health, and a third level of health that indicates a high probability of failure that is higher than that of the second level of health.

10. The computer-implemented method of claim 9, further comprising, as a result of determining that the level of health of the storage device is within the third level of health:

causing data of the storage device to be stored in a second storage device;

causing one or more actions to be performed that disable operations to a storage space within the storage device; and

associating the storage device within the third level of health in the data store to enable the administrative computer system to utilize the storage device for one or more purposes associated with the third level of health.

11. The computer-implemented method of claim 8, wherein the remaining storage space of the storage device further comprises storage space that is not allocated to a user.

12. The computer-implemented method of claim 8, wherein the indicator associated with the storage device further includes at least one of a spin-up time or a seek time associated with the storage device.

13. The computer-implemented method of claim 8, wherein the indicator associated with the storage device further includes an indication of firmware, executed by the storage device, that causes disk degradation.

14. A system, comprising:

one or more processors; and

memory that includes instructions that, as a result of being executed by the one or more processors, cause the system to;

receive information associated with a physical or software performance indicator associated with a storage device;

determine, based at least in part on the information, a health level of the storage device, the health level from a scale of possible health levels, at least one health level of the scale of possible health levels includes an intermediate health level between at least a first health level and a second health level;

associate the health level with the storage devices; and

perform a remedial action based at least in part on the health, where the health level indicates at one or more remedial actions for storage devices associated with the health level.

15. The system of claim 14, wherein the physical performance indicator associated with the storage device is based at least in part on at least one of a latency for a read request, a latency for a write request, a spin-up time, a seek time, a number of attempts to complete an operation, a temperature, or an overall variance in behavior associated with the storage device.

16. The system of claim 14, wherein the software performance indicator associated with the storage device is based at least in part on at least one of a type of software associated with the storage device or a predefined software detail associated with performance of the storage device.

17. The system of claim 14, wherein the first health level indicates that the storage device is healthy, the intermediate health level indicates that the storage device is suspect, and the second health level indicates that the storage device is non-operational.

18. The system of claim 14, wherein the one or more remedial actions indicated by the health level further includes storing data, stored in the storage device, to a second storage device with a health level indicating a lower probability of failure, wherein the first health level has a lower probability of failure than the intermediate health level, and wherein the intermediate health level has a lower probability of failure than the second health level.