System and method for managing latency prediction for a storage system
An information handling system for managing a storage system includes storage and a storage manager. The storage is for storing a concurrency effect model for the storage system. The storage manager obtains a request for a latency prediction for the storage system; obtains media access time estimates for the storage system using a trace that specifies: a series of accesses; and responses, by the storage system, to the series of accesses; obtain concurrency estimates for the trace using the media access time estimates; obtains modifications for the media access time estimates using: the concurrency effect model; and the concurrency estimates; updates the media access time estimates using the modifications to obtain a latency prediction for the storage system; and performs an action set, based on the latency prediction, to manage operation of the storage system.
1. An information handling system for managing a storage system, comprising:
storage for storing a concurrency effect model for the storage system; and
a storage manager programmed to:
obtain a request for a latency prediction for the storage system;
obtain media access time estimates for the storage system using a trace that specifies:
a series of accesses; and
responses, by the storage system, to the series of accesses, wherein the media access time estimates are baseline response times for the series of accesses;
obtain concurrency estimates for the trace using the media access time estimates;
obtain modifications for the media access time estimates using:
the concurrency effect model; and
the concurrency estimates, wherein the modifications are additional duration of time beyond each of the baseline response times;
update the media access time estimates using the modifications to obtain the latency prediction for the storage system; and
perform an action set, based on the latency prediction, to manage operation of the storage system, wherein the action set comprises:
providing the latency prediction for the storage system to an entity;
making a determination that the latency prediction meets a predetermined requirement, wherein the predetermined requirement indicates that the latency prediction is desirable; and
in response to the determination, configuring operation of the storage system based on a configuration associated with the trace.
2. The information handling system of claim 1 , wherein the storage manager is further programmed to:
prior to obtaining the request for a latency prediction:
obtain a second trace and a media configuration profile for the storage system;
obtain second concurrency estimates using the second trace and the media configuration profile;
generate the concurrency effect model using the concurrency estimates, the second trace, and the media configuration profile.
3. The information handling system of claim 2 , wherein the concurrency effect model is generating using a machine learning algorithm that takes, as input, the second trace, second concurrency estimates, and the media configuration profile.
4. The information handling system of claim 1 , wherein the storage system comprises a storage device and a cache for the storage device.
5. The information handling system of claim 1 , wherein the trace was obtained while the storage system was operating based on the configuration.
6. A method for managing a storage system, comprising:
obtaining a request for a latency prediction for the storage system;
obtaining media access time estimates for the storage system using a trace that specifies:
a series of accesses; and
responses, by the storage system, to the series of accesses, wherein the media access time estimates are baseline response times for the series of accesses;
obtaining concurrency estimates for the trace using the media access time estimates;
obtaining modifications for the media access time estimates using:
a concurrency effect model associated with the storage system; and
the concurrency estimates, wherein the modifications are additional duration of time beyond each of the baseline response times;
updating the media access time estimates using the modifications to obtain the latency prediction for the storage system; and
performing an action set, based on the latency prediction, to manage operation of the storage system, wherein the action set comprises:
providing the latency prediction for the storage system to an entity;
making a determination that the latency prediction meets a predetermined requirement, wherein the predetermined requirement indicates that the latency prediction is desirable; and
in response to the determination, configuring operation of the storage system based on a configuration associated with the trace.
7. The method of claim 6 , further comprising:
prior to obtaining the request for a latency prediction:
obtaining a second trace and a media configuration profile for the storage system;
obtaining second concurrency estimates using the second trace and the media configuration profile;
generating the concurrency effect model using the concurrency estimates, the second trace, and the media configuration profile.
8. The method of claim 7 , wherein the concurrency effect model is generating using a machine learning algorithm that takes, as input, the second trace, second concurrency estimates, and the media configuration profile.
9. The method of claim 6 , wherein the storage system comprises a storage device and a cache for the storage device.
10. The method of claim 6 , wherein the trace was obtained while the storage system was operating based on the configuration.
11. A non-transitory computer readable medium comprising computer readable program code, which when executed by a computer processor enables the computer processor to perform a method for managing a storage system, the method comprising:
obtaining a request for a latency prediction for the storage system;
obtaining media access time estimates for the storage system using a trace that specifies:
a series of accesses; and
responses, by the storage system, to the series of accesses, wherein the media access time estimates are baseline response times for the series of accesses;
obtaining concurrency estimates for the trace using the media access time estimates;
obtaining modifications for the media access time estimates using:
a concurrency effect model associated with the storage system; and
the concurrency estimates, wherein the modifications are additional duration of time beyond each of the baseline response times;
updating the media access time estimates using the modifications to obtain the latency prediction for the storage system; and
performing an action set, based on the latency prediction, to manage operation of the storage system, wherein the action set comprises:
providing the latency prediction for the storage system to an entity;
making a determination that the latency prediction meets a predetermined requirement, wherein the predetermined requirement indicates that the latency prediction is desirable; and
in response to the determination, configuring operation of the storage system based on a configuration associated with the trace.
12. The non-transitory computer readable medium of claim 11 , wherein the method further comprises:
prior to obtaining the request for a latency prediction:
obtaining a second trace and a media configuration profile for the storage system;
obtaining second concurrency estimates using the second trace and the media configuration profile;
generating the concurrency effect model using the concurrency estimates, the second trace, and the media configuration profile.
13. The non-transitory computer readable medium of claim 12 , wherein the concurrency effect model is generating using a machine learning algorithm that takes, as input, the second trace, second concurrency estimates, and the media configuration profile.
14. The non-transitory computer readable medium of claim 11 , wherein the storage system comprises a storage device and a cache for the storage device.