Thermal aware workload scheduling
Systems, software, devices, and methods of distributing a workload among available data storage devices in a thermal aware manner are described herein. More specifically, the examples herein discuss distributing the workload among the available data storage devices in a thermal aware manner that optimizes collective IOPs of the data storage devices in an enclosure. The thermal aware distribution of the storage operations is determined by a thermal model that predicts thermal characteristics of the data storage system based on inlet air characteristics of the enclosure, performance characteristics and thermal constraints of the data storage devices, and constraints of the workload.
1. A data storage system, comprising:
multiple data storage devices, each configured to store data;
an enclosure configured to encase and physically support the multiple data storage devices; and
a control system configured to:
model thermal characteristics of the data storage system based on inlet air characteristics of the enclosure, performance characteristics and thermal constraints of the data storage devices, and constraints of a workload;
identify, based on the thermal model, a thermal-aware allocation of the workload among the data storage devices; and
distribute the workload among the data storage devices in accordance with the thermal-aware allocation.
2. A method of distributing a workload among data storage devices encased in an enclosure of a data storage system, the method comprising:
identifying performance characteristics and thermal constraints associated with the data storage devices;
identifying inlet air characteristics of the enclosure;
identifying constraints associated with the workload;
modeling thermal characteristics of the data storage system based on the inlet air characteristics of the enclosure, the performance characteristics and thermal constraints of the data storage devices, and the constraints associated with the workload;
identifying, based on the thermal model, a thermal-aware allocation of the workload among the data storage devices; and
distributing the workload among the data storage devices in accordance with the thermal-aware allocation.
3. The method of claim 2 , wherein the thermal-aware allocation maximizes collective Input/Output Operations Per Second (IOPS) of the multiple data storage devices relative to other allocations of the workload while maintaining the multiple data storage devices within the thermal constraints of the data storage devices.
4. The method of claim 2 , wherein identifying the inlet air characteristics of the enclosure comprises estimating one or more of the inlet air characteristics based on fan drive signal characteristics.
5. The method of claim 2 , wherein:
identifying the inlet air characteristics of the enclosure comprises monitoring air temperature sensors and/or air speed sensors to determine one or more of the inlet air characteristics; and
the inlet air characteristics comprise one or more of air temperature and/or air speed.
6. The method of claim 2 , wherein the performance characteristics of the data storage devices comprise power versus performance operating curves of each of the data storage devices.
7. The method of claim 2 , wherein the thermal constraints of the data storage devices comprise one or more safe operating temperature thresholds.
8. The method of claim 2 , wherein the constraints of the workload comprise a breakdown of read operations and write operations.
9. A computer-readable storage medium having instructions stored thereon which, when executed by one or more processors, cause the one or more processors to:
identify performance characteristics and thermal constraints associated with data storage devices encased in an enclosure of a data storage system;
identify inlet air characteristics of the enclosure;
identify constraints associated with a workload;
model thermal characteristics of the data storage system based on the inlet air characteristics of the enclosure, the performance characteristics and thermal constraints of the data storage devices, and the constraints associated with the workload;
identify, based on the thermal model, a thermal-aware allocation of the workload among the data storage devices; and
distribute the workload in accordance with the thermal-aware allocation.
10. A control system for a data storage system, wherein the data storage system comprises an enclosure and data storage devices encased within the enclosure, the control system comprising:
means for modeling thermal characteristics of the data storage system based on inlet air characteristics of the enclosure, performance characteristics and thermal constraints of the data storage devices, and constraints of a workload;
means for identifying, based on the modeled thermal characteristics, a thermal-aware allocation of the workload among the data storage devices; and
means for distributing the workload among the data storage devices in accordance with the thermal-aware allocation.
11. A data storage assembly comprising:
an enclosure configured to encase and physically support multiple data storage devices; and
a control system configured to:
identify a thermal model for the data storage assembly based at least on inlet air characteristics of the enclosure, performance characteristics and thermal constraints of the data storage devices, and characteristics of a workload;
identify, based on the thermal model, a thermal-aware allocation of the workload among the data storage devices; and
distribute the workload among the data storage devices in accordance with the thermal-aware allocation.
12. The data storage assembly of claim 11 , wherein the thermal-aware allocation maximizes collective Input/Output Operations Per Second (IOPS) of the multiple data storage devices relative to other allocations of the workload while maintaining the multiple data storage devices within the thermal constraints of the data storage devices.
13. The data storage assembly of claim 11 , wherein the control system is further configured to estimate one or more of the inlet air characteristics based on fan drive signal characteristics.
14. The data storage assembly of claim 13 , wherein the fan drive signal characteristics comprise a voltage or a duty cycle.
15. The data storage assembly of claim 11 , wherein:
the control system is further configured to monitor air temperature sensors and/or air speed sensors to determine one or more of the inlet air characteristics, and
the inlet air characteristics comprise one or more of air temperature and/or air speed.
16. The data storage assembly of claim 11 , further comprising:
one or more fans configured to provide airflow within the enclosure;
thermal sensors configured to measure temperatures of each of the multiple data storage devices; and
one or more sensors configured to measure one or more of the inlet air characteristics.
17. The data storage assembly of claim 11 , wherein the performance characteristics of the data storage devices comprise power versus performance operating curves of each of the data storage devices.
18. The data storage assembly of claim 11 , wherein the data storage devices are thermally-coupled within the enclosure.
19. The data storage assembly of claim 11 , wherein the thermal constraints of the data storage devices comprise one or more safe operating temperature thresholds.
20. The data storage assembly of claim 19 , wherein the control system is further configured to:
monitor temperatures of each of the multiple data storage devices to ensure that the data storage devices do not exceed the one or more safe operating temperature thresholds.
21. The data storage assembly of claim 11 , wherein the constraints of the workload comprise a breakdown of read operations and write operations.