CONTROLLING DATA STORAGE INPUT/OUTPUT REQUESTS

• US 20150081948A1
• Filed: 09/13/2013
• Published: 03/19/2015
• Est. Priority Date: 09/13/2013
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Thereska, Eno, Ballani, Hitesh, O'Shea, Gregory, Karagiannis, Thomas, Rowstron, Antony Ian Taylor, Black, Richard John, Talpey, Thomas M
• Granted Patent: US 9,244,615 B2
• Time in Patent Office: Days
• US Class Current: 711/103
• CPC Class Codes: G06F 11/3684 for test design, e.g. generating new test cases
• CPC Class Codes: G06F 3/061 Improving I/O performance
• CPC Class Codes: G06F 3/0631 by allocating resources to storage systems
• CPC Class Codes: G06F 3/0673 Single storage device
• CPC Class Codes: H04L 45/20 Hop count for routing purposes, e.g. TTL
• CPC Class Codes: H04L 45/50 using label swapping, e.g. multi-protocol label switch [MPLS]
• CPC Class Codes: H04L 47/10 Flow control; Congestion control
• CPC Class Codes: H04L 47/20 Traffic policing
• CPC Class Codes: H04L 47/22 Traffic shaping
• CPC Class Codes: H04L 47/26 using explicit feedback to the source, e.g. choke packets
• CPC Class Codes: H04L 47/52 by attributing bandwidth to queues
• CPC Class Codes: H04L 47/522 Dynamic queue service slot or variable bandwidth allocation
• CPC Class Codes: H04L 49/90 Buffering arrangements

Abstract

Controlling data storage input/output requests is described, for example, to apply a policy to an end-to-end flow of data input/output requests between at least one computing entity and at least one store. In various examples a plurality of queues are configured at one or more stages of the end-to-end flow and controlled to adhere to a policy. In examples, each stage has a control interface enabling it to receive and execute control instructions from a controller which may be centralized or distributed. For example, the control instructions comprise queuing rules and/or queue configurations. In various examples queues and queuing rules are dynamically created and revised according to feedback about any of: flow behavior, changes in policy, changes in infrastructure or other factors. In examples, high level identifiers of the flow endpoints are resolved, on a per stage basis, to low level identifiers suitable for use by the stage.

Claims

1. A computer-implemented method at a data storage system comprising:

accessing, at a processor, a policy comprising one or more criteria to be applied to a flow of data storage input/output requests between endpoints of the flow, the endpoints comprising at least one computing entity and at least one store;

sending control instructions to at least one stage of the flow of data, the stage being a component of the data storage system through which the flow passes and which is configured to interpret the control instructions, the control instructions arranged to control a plurality of queues at the at least one stage in order to treat the flow of data storage input/output requests such that the policy is met.

2. The method as claimed in claim 1 wherein the flow is between three or more endpoints.

3. The method as claimed in claim 1 wherein the flow of data storage input/output requests uses resources which are shared, at least in part, with one or more other flows.

4. The method as claimed in claim 1 wherein accessing the policy comprises accessing high level identifiers of the at least one computing entity and the at least one store, where high level identifiers are identifiers which are not understood by the at least one stage, and where the method comprises resolving the high level identifiers into low level identifiers which are understood by the at least one stage.

5. The method as claimed in claim 1 carried out at a centralized controller in the data storage system, the data storage system having a plurality of flows of data storage input/output requests between a plurality of computing entities and a plurality of stores, each flow having at least one stage.

6. The method as claimed in claim 5 comprising maintaining topology of the plurality of computing entities and the plurality of stores and information about stages of the data storage system.

7. The method as claimed in claim 1 wherein the control instructions are arranged to create a queuing rule specifying which types of data storage input/output requests are to be placed into a specified queue at a stage.

8. The method as claimed in claim 1 wherein the control instructions are arranged to configure service characteristics of a queue.

9. The method as claimed in claim 1 wherein the control instructions are arranged to configure routing characteristics of a queue.

10. The method as claimed in claim 1 wherein the stage is at any of:

the at least one computing entity, the at least one store, a switch present in an interconnection fabric between the at least one computing entity and the at least one store.

11. The method as claimed in claim 1 comprising selecting the at least one stage from a plurality of available stages according to criteria such as efficiency of policy enforcement.

12. The method as claimed in claim 1 comprising obtaining feedback about the flow of data storage input/output requests and modifying the control instructions on the basis of the feedback.

13. A computer-implemented method at a component of a data storage system comprising:

receiving, at a processor, a flow of data storage input/output requests between at least one computing entity and at least one store;

receiving control instructions;

configuring one or more queues according to the control instructions;

allocating the data storage input/output requests to the queues according to queuing rules which are part of the control instructions and servicing the queues.

14. The method of claim 13 comprising allocating the data storage input/output requests to the queues by inspecting headers of the data storage input output requests.

15. The method of claim 13 comprising, receiving a data storage input/output request which cannot be allocated to the one or more queues according to the queuing rules;

and sending a request to a controller requesting a new queue.

16. The method of claim 13 comprising receiving a queue from a controller;

installing the queue at the component and receiving at least one queuing rule for use with the installed queue.

17. A data storage system component comprising:

a processor arranged to transmit a flow of data storage input/output requests between at least one computing entity and at least one store through a plurality of queues at the processor;

an interface arranged to receive control instructions from a controller of the data storage system;

the processor arranged to configure the queues according to the control instructions and to allocate the data storage input/output requests to the queues according to queuing rules which are part of the control instructions.

18. The data storage system component of claim 17, the processor arranged to allocate the data storage input/output requests to the queues by inspecting headers of the data storage input output requests.

19. The data storage system component of claim 17 wherein the processor is arranged to receive a queue from the controller, install the queue at the component and to receive at least one queuing rule for use with the installed queue.

20. The data storage system component of claim 17 being at least partially implemented using hardware logic selected from any one or more of:

a field-programmable gate array, a program-specific integrated circuit, a program-specific standard product, a system-on-a-chip, a complex programmable logic device