Systems and methods for optimizing data storage among a plurality of storage drives

Patent metadata

Field	Value
Publication number	US-10191667-B2
Application number	US-201615017071-A
Country	US
Kind code	B2
Filing date	Feb 5, 2016
Priority date	Oct 10, 2010
Publication date	Jan 29, 2019
Grant date	Jan 29, 2019

Abstract

A data storage device is presented that includes an interface system configured to communicate over an aggregated host link comprising Peripheral Component Interconnect Express (PCIe) interfaces to receive one or more write operations for storage of data by the data storage device. The data storage device includes plurality solid state drives (SSDs) each comprising drive PCIe interfaces coupled to the interface system. The interface system is configured to process the one or more write operations against storage allocation information to parallelize the data among the SSDs and transfer portions of the parallelized data to associated SSDs over corresponding drive PCIe interfaces.

First claim

What is claimed is:

1. A data storage device, comprising:
an interface system configured to communicate over an aggregated host interface to receive one or more write operations for storage of data by the data storage device, wherein the aggregated host interface comprises a selected quantity of Peripheral Component Interconnect Express (PCIe) interface lanes combined to form the aggregated host interface;
a plurality solid state drives (SSDs) each comprising an associated drive PCIe interface coupled to the interface system; and
the interface system configured to process the one or more write operations against storage allocation information to parallelize the data into individual portions for storage among more than one of the SSDs and transfer corresponding portions of the parallelized data to associated SSDs over corresponding drive PCIe interfaces.

2. The data storage device of claim 1 , comprising:
the SSDs configured to receive the corresponding portions of the data and responsively store the corresponding portions on associated solid state storage media.

3. The data storage device of claim 1 , comprising:
the interface system configured to determine ones of the SSDs to apportion the data among based at least on processing the data against the storage allocation information to determine which of the SSDs will store the data.

4. The data storage device of claim 1 , comprising:
the interface system configured to vary the selected quantity of PCIe interface lanes combined into the aggregated host interface based at least on utilization levels associated with the data storage device.

5. The data storage device of claim 1 , comprising:
the interface system configured to communicate with the host system over the aggregated host interface to receive one or more read operations for retrieving data previously stored by the data storage device;
the interface system configured to process at least read addresses indicated by the one or more read operations against storage locations indicated by the storage allocation information to determine ones of the SSDs to handle the read operations, and transfer the read operations over the drive PCIe interfaces to associated ones of the SSDs;
the SSDs configured to receive associated ones of the read operations, and in response, retrieve read data from the associated solid state storage media and transfer the read data to the interface system for transfer to the host system over the aggregated host interface.

6. The data storage system of claim 5 , comprising:
based at least on the read data having been parallelized during storage over the more than one of the SSDs, the interface system configured to process the read data received from the more than one of the SSDs against the storage allocation information to de-parallelize the read data into merged read data, and transfer the merged read data to the host system over the aggregated host interface.

7. The data storage device of claim 1 , comprising:
the interface system configured to parallelize the data for storage among the more than one of the SSDs by at least forming the data into at least partially redundant data portions for storage as the portions by an associated SSDs.

8. The data storage device of claim 1 , further comprising:
a configuration system configured to receive a storage configuration over the aggregated host interface, wherein the storage configuration indicates at least a quantity and arrangement of virtual volumes;
the interface system configured to apply the storage configuration to subdivide the SSDs into a plurality of virtual volumes;
the interface system configured to present the plurality of virtual volumes as separate drive volumes to a host system over the aggregated host interface;
wherein storage operations received over the aggregated host interface comprise indicators of at least one of the drive volumes for handling of associated data; and
the interface system configured to parallelize the associated data for storage among a plurality of the SSDs which comprise indicated ones of the virtual volumes.

9. The data storage device of claim 1 , further comprising:
a power control system configured to receive backup power from a backup power source, and detect an interruption in a primary power source and in response provide the backup power to the data storage device; and
the interface system configured to perform a soft power down operation to commit at least pending write instructions and associated write data to at least one of the SSDs in response to the interruption in the primary power source.

10. The data storage device of claim 9 , comprising:
the interface system configured to commit the storage allocation information to non-volatile memory responsive to the soft power down operation.

11. A method of operating a data storage device, the method comprising:
in an interface system, communicating over an aggregated host interface to receive one or more write operations for storage of data by the data storage device on ones of a plurality of solid state drives (SSDs) each comprising an associated drive Peripheral Component Interconnect Express (PCIe) interface coupled to the interface system, wherein the aggregated host interface comprises a selected quantity of PCIe interface lanes combined to form the aggregated host interface; and
in the interface system, processing the one or more write operations against storage allocation information to parallelize the data into individual portions for storage among more than one of the SSDs and transferring corresponding portions of the parallelized data to associated SSDs over corresponding drive PCIe interfaces.

12. The method of claim 11 , further comprising:
in the SSDs, receiving the corresponding portions of the data and responsively store the corresponding portions on associated solid state storage media.

13. The method of claim 11 , further comprising:
in the interface system, determining ones of the SSDs to apportion the data among based at least on processing the data against the storage allocation information to determine which of the SSDs will store the data.

14. The method of claim 11 , further comprising:
in the interface system, varying the selected quantity of PCIe interface lanes combined into the aggregated host interface based at least on utilization levels associated with the data storage device.

15. The method of claim 11 , further comprising:
in the interface system, communicating with the host system over the aggregated host interface to receive one or more read operations for retrieving data previously stored by the data storage device;
in the interface system, processing at least read addresses indicated by the one or more read operations against storage locations indicated by the storage allocation information to determine ones of the SSDs to handle the read operations, and transferring the read operations over the drive PCIe interfaces to associated ones of the SSDs;
in the SSDs, receiving associated ones of the read operations, and in response, retrieving read data from the associated solid state storage media and transferring the read data to the interface system for transfer to the host system over the aggregated host interface.

16. The method of claim 15 , further comprising:
based at least on the read data having been parallelized during storage over the more than one of the SSDs, in the interface system, processing the read data received from the more than one of the SSDs against the storage allocation information to de-parallelize the read data into merged read data,

Assignees

• Liqid Inc

Inventors

• Breakstone Jason
• Gupta Alok
• Desai Himanshu
• Campos Angelo

Classifications

• G06F13/4022

  using switching circuits, e.g. switching matrix, connection or expansion network (G06F13/4009 takes precedence) · CPC title

• G06F2213/0026

  PCI express · CPC title

• G06F3/0665

  at area level, e.g. provisioning of virtual or logical volumes · CPC title

• G06F3/0604

  Improving or facilitating administration, e.g. storage management · CPC title

• G06F3/0613 Primary

  in relation to throughput · CPC title