A data storage system and associated method are provided wherein a policy engine continuously collects qualitative information about a network load to the data storage system in order to dynamically characterize the load and continuously correlates the load characterization to the content of a comma
A data storage system and associated method are provided wherein a policy engine continuously collects qualitative information about a network load to the data storage system in order to dynamically characterize the load and continuously correlates the load characterization to the content of a command queue of transfer requests for writeback commands and host read commands, selectively limiting the content with respect to writeback commands to only those transfer requests for writeback data that are selected on a physical zone basis of a plurality of predefined physical zones of a storage media.
대표청구항▼
1. A data storage system connectable to a computer network, the system comprising: a data storage media having a plurality of zones, each zone allocated to store a book of data; anda nontransitory, tangible computer memory having computer instructions configured to characterize a network load of tra
1. A data storage system connectable to a computer network, the system comprising: a data storage media having a plurality of zones, each zone allocated to store a book of data; anda nontransitory, tangible computer memory having computer instructions configured to characterize a network load of transfer requests from a host to the data storage system, to determine whether the network load has a uniform or a non-uniform distribution based on the characterizing of the network load, to populate a command queue with transfer requests to a plurality of books when the network load is determined to have a substantially uniform distribution of the data storage media, else populate the command queue with only the transfer requests for a selected book when the network load is determined to have a substantially non-uniform distribution of the data storage media; and to issue the transfer requests of the command queue by transferring data between the host and the respective books of the data storage system. 2. The data storage system of claim 1, wherein the transfer requests comprise writeback requests and host read commands. 3. The data storage system of claim 1, wherein the command queue is populated with transfer requests to a plurality of books for a predefined interval during which the network load is characterized. 4. The data storage system of claim 1, wherein characterizing the network load comprises characterizing the network load in terms of a ratio of writeback requests to host read commands. 5. The data storage system of claim 1, wherein characterizing the network load comprises characterizing the network load in terms of a rate of received transfer requests. 6. The data storage system of claim 1, wherein characterizing the network load comprises characterizing the network load in terms of a ratio of rate sensitive transfer requests to latency sensitive transfer requests. 7. The data storage system of claim 6, wherein the rate sensitive transfer requests are for writeback cache commands and the latency sensitive transfer requests are for at least one of read commands and write through cache commands. 8. The data storage system of claim 1, wherein characterizing the network load comprises characterizing the network load in terms of a bandwidth associated with each transfer request. 9. The data storage system of claim 1, wherein each zone is an annular portion of the data storage media. 10. The data storage system of claim 1, comprising an array of data storage media allocated to define a redundant array of independent drives (RAID), wherein respective zones are allocated stripes of data storage capacity for two or more of the data storage media. 11. The data storage system of claim 1, wherein the computer instructions further comprise instructions to selectively issue transfer requests in the command queue to a command profile and to define rules responsive to the network load characterization and responsive to predefined goals that govern issuing the selected transfer requests to the command profile. 12. The data storage system of claim 11, wherein characterizing the network load comprises characterizing the network load in terms of a ratio of rate sensitive transfer requests to latency sensitive transfer requests and wherein the instructions further comprise selectively populating the command profile responsive to a respective predetermined maximum latency of the transfer requests. 13. The data storage system of claim 11, wherein the computer instructions further comprise instructions to selectively populate the command profile responsive to a predetermined prioritization of different LUN classes. 14. The data storage system of claim 1, further comprising a circuit configured to execute the computer instructions. 15. A method comprising: characterizing a network load of transfer requests from a host to a data storage system, the data storage system having a data storage media with a plurality of zones, each zone allocated to store a book of data;determining whether the network load has a uniform or a non-uniform distribution based on the characterizing of the network load;populating a command queue with a portion of the transfer requests of the network load, the portion of the transfer requests in the command queue being associated with a plurality of books of the data storage media responsive to the network load determined to have a substantially uniform distribution, the portion of the transfer requests in the command queue being associated with only a selected book of the data storage media responsive to the network load determined to have a substantially non-uniform distribution; andissuing the portion of the transfer requests of the command queue by transferring data between the host and the respective books of the data storage media. 16. The method of claim 15, wherein the characterizing step comprises characterizing the network load in terms of a ratio of rate sensitive transfer requests to latency sensitive transfer requests. 17. The method of claim 16, wherein the characterizing step comprises identifying write back cache commands as rate sensitive transfer requests and identifying at least one of read commands and writethrough cache commands as being latency sensitive transfer requests. 18. The method of claim 15, further comprising striping a plurality of the books across a respective plurality of the data storage media allocated to define a redundant array of independent drives (RAID). 19. A data storage system comprising: a data storage media having a plurality of zones, each zone allocated to store a book of data;a fabric interface coupled to a host computer system;a command queue storing a sequence of transfer requests from the host computer system; anda cache manager circuit coupled to the host interface and to the command queue, the cache manager circuit configured to characterize a network load of transfer requests from the host computer system to the data storage system, to determine whether the network load has a uniform or a non-uniform distribution based on the characterizing of the network load, to populate the command queue with transfer requests to a plurality of books when the network load is a substantially uniform distribution of the data storage media, and to populate the command queue with only the transfer requests for a selected book when the network load is a substantially non-uniform distribution of the data storage media; anda device interface coupled to the data storage media, the device interface configured to flush the command queue by transferring data between the host and the respective books of the data storage system. 20. The data storage system of claim 19, wherein the cache manager populates the command queue with transfer requests to a plurality of books for a predefined interval during which the network load is characterized.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (8)
Codilian, Raffi; Webb, Steven L., Adaptively modifying a read caching algorithm based upon the detection of a vibration state within a rotating media storage device.
Matsunami Naoto,JPX ; Yoshida Minoru,JPX ; Miyazawa Shoichi,JPX ; Oeda Takashi,JPX ; Honda Kiyoshi,JPX ; Ohno Shuji,JPX, Array disk controller for grouping host commands into a single virtual host command.
McKnight, Gregory Joseph; Riedle, Linda Ann; Stephan, Charles Thorpe, Method and system for improving raid controller performance through adaptive write back/write through caching.
Noel S. Otterness ; William A. Brant ; Keith E. Short ; Joseph G. Skazinski, System, apparatus and method for multi-level cache in a multi-processor/multi-controller environment.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.