Cloud black box for cloud infrastructure

Patent metadata

Field	Value
Publication number	US-9307015-B1
Application number	US-201314141643-A
Country	US
Kind code	B1
Filing date	Dec 27, 2013
Priority date	Dec 27, 2013
Publication date	Apr 5, 2016
Grant date	Apr 5, 2016

Abstract

A cloud black box (CBB) subsystem in a cloud computing infrastructure includes CBB storage and computer processing circuitry executing a CBB application having first and second operating modes. In a depository mode information messages are continually received from hardware computing devices during normal operation and device information from the messages is stored into the CBB storage. The information messages are generated by CBB agents executing on the hardware computing devices, which continually collect the device information and generate the information messages according to a common information transfer protocol. In a retrieval mode, device information in the CBB storage is provided to a requestor such as a data analysis application, which may be part of or external to the CBB subsystem. The CBB subsystem operates independently and remains available upon failure of hardware or software components in the cloud infrastructure, providing a centralized source of information for diagnosis or other analysis.

First claim

What is claimed is:

1. A cloud black box subsystem for use in a cloud computing infrastructure having a set of network-connected hardware computing devices including host computers, storage arrays and network switches, the host computers being operative to execute one or more applications of a cloud infrastructure user, the applications engaging in application data storage accesses with the storage arrays and engaging in application network data transfers via the network switches, comprising:
cloud black box storage; and
computer processing circuitry operative to execute a cloud black box application having first and second operating modes, the first operating mode being a depository mode in which information messages are continually received from the hardware computing devices during normal operation of the cloud computing infrastructure and device information from the information messages is stored into the cloud black box storage, the information messages generated by respective cloud black box agents executing on the hardware computing devices and continually operative to collect the device information during operation of the respective hardware computing devices and to generate the information messages containing the device information according to a common information transfer protocol, the second operating mode being a retrieval mode in which the device information in the cloud black box storage is provided to a data analysis application, the cloud black box subsystem operating independently of the applications and hardware computing devices so as to be functional in the retrieval mode in an event of failure or disruption of any of the applications and hardware computing devices.

2. A cloud black box subsystem according to claim 1 , realized as a distinct physical computing system having one or more physical enclosures housing physical computing elements dedicated to the cloud black box subsystem.

3. A cloud black box subsystem according to claim 2 , wherein the cloud black box subsystem is realized as one or more network attached storage (NAS) storage nodes mounted to one or more physical cloud black box server computers.

4. A cloud black box subsystem according to claim 2 , wherein the cloud black box subsystem is realized as one or more cloud black box host computers having local physical storage.

5. A cloud black box subsystem according to claim 1 , realized as in a virtualized manner including a virtual service executing in a computing cloud and having storage in the computing cloud.

6. A cloud black box subsystem according to claim 1 , wherein the common information transfer protocol is defined by an application programming interface (API) exported by the cloud black box subsystem that provides functions of (i) logging a message, (ii) depositing a blob of data, and (iii) raising an alert.

7. A cloud black box subsystem according to claim 6 , wherein the API reflects predetermined guidelines for logging device information.

8. A cloud black box subsystem according to claim 1 , operative in the depository mode to apply filtering to the device information received in the information messages to improve efficiency of storage of the device information.

9. A cloud black box subsystem according to claim 1 , wherein the cloud black box storage employs tiered storage having at least two levels, a first level providing fast access to current and frequently accessed device information, and a second level providing high-capacity storage of older and infrequently accessed device information.

10. A cloud black box subsystem according to claim 1 ,
wherein the data analysis application is executed remotely in a separate computer system and the cloud black box subsystem exports a retrieval application programming interface (API) to provide access to the stored device information by the remote data analysis application.

11. A cloud black box subsystem according to claim 1 , wherein the data analysis application is executed locally in the cloud black box subsystem and exports a user interface to the cloud black box users.

12. A non-transitory computer-readable medium storing computer program instructions, the instructions being executable as a cloud black box application by a set of one or more computers to cause the computers to form, along with cloud black box storage, a cloud black box subsystem for use in a cloud computing infrastructure having a set of network-connected hardware computing devices including host computers, storage arrays and network switches, the host computers being operative to execute one or more applications of a cloud infrastructure user, the applications engaging in application data storage accesses with the storage arrays and engaging in application network data transfers via the network switches, the cloud black box subsystem performing a method including:
a first operating mode being a depository mode in which information messages are continually received from the hardware computing devices during normal operation of the cloud computing infrastructure and device information from the information messages is stored into the cloud black box storage, the information messages generated by respective cloud black box agents executing on the hardware computing devices and continually operative to collect the device information during operation of the respective hardware computing devices and to generate the information messages containing the device information according to a common information transfer protocol; and
a second operating mode being a retrieval mode in which the device information in the cloud black box storage is provided to a data analysis application, the cloud black box subsystem operating independently of the applications and hardware computing devices so as to be functional in the retrieval mode in an event of failure or disruption of any of the applications and hardware computing devices.

13. A non-transitory computer-readable medium according to claim 12 , wherein the common information transfer protocol is defined by an application programming interface (API) exported by the cloud black box subsystem that provides functions of (i) logging a message, (ii) depositing a blob of data, and (iii) raising an alert.

14. A non-transitory computer-readable medium according to claim 13 , wherein the API reflects predetermined guidelines for logging device information.

15. A non-transitory computer-readable medium according to claim 12 , wherein the depository mode includes applying filtering to the device information received in the information messages to improve efficiency of storage of the device information.

16. A non-transitory computer-readable medium according to claim 12 , wherein the cloud black box storage employs tiered storage having at least two levels, a first level providing fast access to current and frequently accessed device information, and a second level providing high-capacity storage of older and infrequently accessed device information.

17. A non-transitory computer-readable medium according to claim 12 , wherein the data analysis application is executed remotely in a separate computer system and the cloud black box subsystem exports a retrieval API to provide access to the stored device information by the remote data analysis application.

18. A non-transitory computer-readable medium according to claim 12 , wherein the data analysis application is executed locally in the cloud black box subsystem and exports a user interface to the cloud black box users.

19. A cloud computing infrastructure, comprising:

Assignees

• Emc Corp

Inventors

• Bauer Andreas L
• Tang Xuan
• Parenti David
• Regan Andrew

Classifications

• G06F3/0685

  Hybrid storage combining heterogeneous device types, e.g. hierarchical storage, hybrid arrays · CPC title

• G06F3/0649

  Lifecycle management · CPC title

• G06F3/0608

  Saving storage space on storage systems · CPC title

• G06F3/061

  Improving I/O performance · CPC title

• G06F3/067

  Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS] · CPC title