Storage system and method for stream management in a multi-host virtualized storage system
A storage system and method for stream management in a multi-host virtualized storage system are provided. In one embodiment, a method for stream management is provided that is performed in a storage system in communication with a host comprising a plurality of virtual hosts. The method comprises: receiving, from the host, identification of each virtual host of the plurality of virtual hosts; analyzing usage history of each virtual host of the plurality of virtual hosts; and assigning streams to a subset of the plurality of virtual hosts based on the usage history, wherein a maximum number of streams assignable by the storage system is less than a total number of virtual hosts in the plurality of virtual hosts. Other embodiments are provided.
1. A method for stream management, the method comprising:
performing the following in a storage system in communication with a host comprising a plurality of virtual hosts:
receiving, from the host, identification of each virtual host of the plurality of virtual hosts;
analyzing hit rates of read look-ahead algorithms used for each virtual host of the plurality of virtual hosts;
assigning streams to a subset of the plurality of virtual hosts based on the hit rates, wherein:
a maximum number of streams assignable by the storage system is less than a total number of virtual hosts in the plurality of virtual hosts;
a remainder of the subset of the plurality of virtual hosts are not assigned a stream; and
by not being assigned a stream, the remainder of the subset of the plurality of virtual hosts do not have access to a memory of the storage system;
separately accumulating data from each of the subset of the plurality of virtual hosts; and
writing the separately-accumulated data from each of the subset of the plurality of virtual hosts into a submission queue such that there is no interleaving of data between different virtual hosts of the subset of the plurality of virtual hosts;
wherein assigning streams based on the hit rates increases the hit rates of the read look-ahead algorithms.
2. The method of claim 1 , wherein assigning streams comprises assigning multiple streams to a single virtual host.
3. The method of claim 1 , wherein the hit rates are analyzed as a background operation.
4. The method of claim 1 , wherein the hit rates are analyzed in response to the total number of virtual hosts exceeding a threshold.
5. The method of claim 1 , wherein the hit rates are analyzed after an elapsed period of time.
6. The method of claim 1 , wherein the hit rates are analyzed using a machine-learning algorithm.
7. The method of claim 1 , further comprising assigning a stream identifier to a virtual host.
8. The method of claim 1 , further comprising receiving priority information for the plurality of virtual hosts, and wherein the streams are assigned at least in part based on the priority information.
9. The method of claim 1 , wherein the memory comprises a three-dimensional memory.
10. The method of claim 1 , wherein the storage system is removably connectable to the host.
11. The method of claim 1 , wherein the storage system is embedded in the host.
12. A storage system comprising:
a memory; and
a controller configured to:
receive a plurality of virtual host identifiers;
analyze hit rates of read look-ahead algorithms associated with each virtual host identifier of the plurality of the virtual host identifiers;
assign stream identifiers to a subset of the plurality of virtual host identifiers based on the hit rates, wherein:
a maximum number of stream identifiers assignable by the storage system is less than a total number of virtual host identifiers in the plurality of virtual host identifiers;
a remainder of the subset of the plurality of virtual host identifiers are not assigned a stream identifier; and
by not being assigned a stream identifier, virtual hosts associated with the remainder of the subset of the plurality of virtual host identifiers do not have access to the memory of the storage system;
separately accumulate data from each of virtual host associated with the subset of the plurality of virtual host identifiers; and
write the separately-accumulated data from each virtual host associated with the subset of the plurality of virtual host identifiers into a submission queue such that there is no interleaving of data between different virtual hosts;
wherein assigning stream identifiers based on the hit rates increases the hit rates of the read look-ahead algorithms.
13. The storage system of claim 12 , wherein the memory comprises a three-dimensional memory.
14. A storage system comprising:
a memory;
means for assigning streams to a subset of a plurality of virtual hosts based on hit rates of read look-ahead algorithms used for each of the plurality of virtual hosts, wherein:
a maximum number of streams assignable by the storage system is less than a total number of virtual hosts in the plurality of virtual hosts;
a remainder of the subset of the plurality of virtual hosts are not assigned a stream; and
by not being assigned a stream, the remainder of the subset of the plurality of virtual hosts do not have access to the memory of the storage system;
means for separately accumulating data from each of the subset of the plurality of virtual hosts; and
means for writing the separately-accumulated data from each of the subset of the plurality of virtual hosts into a submission queue such that there is no interleaving of data between different virtual hosts of the subset of the plurality of virtual hosts;
wherein assigning streams based on the hit rates increases the hit rates of the read look-ahead algorithms.
15. The method of claim 1 , further comprising analyzing usage history of each virtual host of the plurality of virtual hosts.
16. The storage system of claim 12 , wherein the controller is further configured to analyze usage history of each virtual host of the plurality of virtual hosts.
17. The storage system of claim 12 , wherein the controller is further configured to analyze the hit rates as a background operation.
18. The storage system of claim 12 , wherein the controller is further configured to analyze the hit rates in response to the total number of virtual host identifiers exceeding a threshold.
19. The storage system of claim 12 , wherein the controller is further configured to analyze the hit rates after an elapsed period of time.
20. The storage system of claim 12 , wherein the controller is further configured to analyze the hit rates using a machine-learning algorithm.