Storage media scrubber
Aspects of the present disclosure relate to detecting and correcting a storage array's drive sectors to prevent data loss. In embodiments, a storage array's telemetry information is monitored. Further, one or more state or operational metrics of the storage array's storage drives are measured. Additionally, each storage drive is scrubbed based on each drive's relative scrubbing priority defined by the telemetry information and each storage drive's state or operation metrics.
1. A method comprising:
monitoring a storage array's telemetry information;
measuring one or more state or operational metrics corresponding to each sector of each of the storage array's storage drives;
determining a degree of unexpected activity density for each sector of each storage drive using the telemetry information and each drive's state or operation metrics, wherein the degree of unexpected activity density relates to a local outlier factor (LOF) corresponding to a snapshot collection window of each storage drive;
comparing each storage drive's activity level and corresponding activity density with respect to their neighboring storage drives' activity levels and related activity densities;
scrubbing each storage drive based on each sector's degree of unexpected activity density; and
controlling an allocation of the storage array's computing resources to disk scrubbing operations based on input/output (IO) operations per second (IOPS) of each IO workload the storage array receives.
2. The method of claim 1 , further comprising:
monitoring one or more input/output (IO) workloads received by the storage array;
identifying one or more characteristics of each workload's IO operations; and
reading prior written data from each storage drive's storage media to detect any errors during a read operation.
3. The method of claim 2 , wherein the one or more characteristics include at least one or more of an IO size, IO type, IO service level, destination target device, data types related to each IO operation, address spaces, IO densities, and response time requirements.
4. The method of claim 1 , wherein the state or operational metrics correspond to one or more of each storage drive's capacity, performance capabilities, idle or operating temperatures, powered on/off conditions, errors, and IO characteristics of IO operations associated with each storage drive, amongst other storage drive metrics.
5. The method of claim 4 , further comprising:
determining each storage drive's performance by analyzing their respective state, operational metrics, received IO sizes, amount and frequency of received IO data read/write requests, and response times, amongst other related performance parameters.
6. The method of claim 1 , further comprising establishing one or more data collection time intervals to collect and analyze the telemetry, state, and operational information/metrics.
7. The method of claim 6 , wherein each data collection time interval is based on patterns identified from historical or current IO workloads received by the storage array.
8. The method of claim 6 , further comprising determining each storage drive's relative scrubbing priority within each data collection time interval.
9. The method of claim 8 , further comprising determining each drive's relative scrubbing priority from each storage drive's respective local outlier factor (LOF), wherein each storage drive's LOF is defined by the drive's relative state or operational metrics.
10. The method of claim 9 , further comprising:
aggregating each drive's state or operational metrics;
generating one or more outlier models from the aggregated state or operational metrics using one or more machine learning (ML) techniques; and
assigning each storage drive with a drive's relative scrubbing priority using the one or more outlier models.
11. An apparatus comprising a processor and memory configured to:
monitor a storage array's telemetry information;
measure one or more state or operational metrics corresponding to each sector of each of the storage array's storage drives;
determine a degree of unexpected activity density for each sector of each storage drive using the telemetry information and each drive's state or operation metrics, wherein the degree of unexpected activity density relates to a local outlier factor (LOF) corresponding to a snapshot collection window of each storage drive;
compare each storage drive's activity level and corresponding activity density with respect to their neighboring storage drives' activity levels and related activity densities;
scrub each storage drive based on each sector's degree of unexpected activity density; and
control an allocation of the storage array's computing resources to disk scrubbing operations based on input/output (IO) operations per second (IOPS) of each IO workload the storage array receives.
12. The apparatus of claim 11 , further configured to:
monitor one or more input/output (IO) workloads received by the storage array;
identify one or more characteristics of each workload's IO operations; and
read prior written data from each storage drive's storage media to detect any errors during a read operation.
13. The apparatus of claim 12 , wherein the one or more characteristics include at least one or more of an IO size, IO type, IO service level, destination target device, data types related to each IO operation, address spaces, IO densities, and response time requirements.
14. The apparatus of claim 11 , wherein the state or operational metrics correspond to one or more of each storage drive's capacity, performance capabilities, idle or operating temperatures, powered on/off conditions, errors, and IO characteristics of IO operations associated with each storage drive, amongst other storage drive metrics.
15. The apparatus of claim 14 , further configured to:
analyze each storage drive's respective state, operational metrics, received IO sizes, amount and frequency of received IO data read/write requests, and response times, amongst other related performance parameters to determine their respective drive performances.
16. The apparatus of claim 11 , further configured to establish one or more data collection time intervals to collect and analyze the telemetry, state, and operational information/metrics.
17. The apparatus of claim 16 , wherein each data collection time interval is based on patterns identified from historical or current IO workloads received by the storage array.
18. The apparatus of claim 16 , further configured to determine each storage drive's relative scrubbing priority within each data collection time interval.
19. The apparatus of claim 18 , further configured to determine each drive's relative scrubbing priority from each storage drive's respective local outlier factor (LOF), wherein each storage drive's LOF is defined by each drive's relative state or operational metrics.
20. The apparatus of claim 19 , further configured to:
aggregate each drive's state or operational metrics;
generate one or more outlier models from the aggregated state or operational metrics using one or more machine learning (ML) techniques; and
assign each storage drive with a drive's relative scrubbing priority using the one or more outlier models.