Method and apparatus for providing at risk information in a cloud computing system having redundancy

• US 9,372,775 B1
• Filed: 09/11/2014
• Issued: 06/21/2016
• Est. Priority Date: 06/28/2012
• Status: Active Grant

• Current Assignee: Emc IP Holding Company LLC (Dell Technologies Inc.)
• Original Assignee: EMC Corporation (Dell Technologies Inc.)
• Inventors: Lam, Cheuk
• Primary Examiner(s): Duncan, Marc
• Application Number:
  US14/483,249
• Time in Patent Office: 649 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 11/008 Reliability or availability analysis
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems
• CPC Class Codes: G06F 11/1666 where the redundant component is memory or memory area
• CPC Class Codes: G06F 11/3409 for performance assessment

Abstract

Methods and apparatus for generating at risk probabilities for a pre-integrated cloud computing system. In one embodiment, a system determines a revised overall risk index after at least two component failures in at least two of the compute domain, storage domain, and storage paths to assist a user in selecting a first one of the at least two component failures to fix first.

Claims

1. A method, comprising:

in a pre-integrated cloud computing system having a compute domain, a network domain, a storage domain, a virtual layer and a management layer, to provide services, the cloud storage system providing redundancy,

determining an at risk probability in the compute domain and a first risk index for the compute domain from the at risk probability in the compute domain;

determining an at risk probability for storage volumes in the storage domain and a second risk index for the storage domain from the at risk probability in the storage domain;

determining an at risk probability for storage paths to the storage domain from the compute domain and a third risk index for the storage paths from the at risk probability of the storage paths;

determining an overall at risk probability for a first one of the services using system resources and an overall risk index for the first one of the services from the overall risk probability; and

determining a revised overall risk index after at least two component failures in at least two of the compute domain, storage domain, and storage paths to assist a user in selecting a first one of the at least two component failures to fix first.

2. The method according to claim 1, further including using a logarithm to generate the overall risk index.

3. The method according to claim 1, wherein the at risk probability for the compute domain is derived from a number of blades used by the first one of the services, a failover rate relating to a number of servers that can fail without putting a compute cluster service into an overcommitted state, an annual failure rate for the blades used by the first one of the services, and an amount of time to replace one of the blades used by the first one of the services.

4. The method according to claim 1, wherein the at risk probability for the storage volumes is derived from a number of storage volumes used by the first one of the services, a number of disks in the storage volumes, a number of disks that can fail in the storage volumes without data loss, an annual disk failure rate, and an amount of time needed to convert a hot spare disk to a contributing disk of the storage volume.

5. The method according to claim 1, wherein the at risk probability for the storage paths is derived from a number of storage volumes used by the first one of the services, a number of blades used by the first one of the services, a number of cables used by the storage volumes, an annual failure rate for the cables, and an amount of time to replace a failed one of the cables.

6. An article, comprising:

a non-transitory computer readable medium containing stored instructions that enable a machine to perform, in a pre-integrated cloud computing system having a compute domain, a network domain, a storage domain, a virtual layer and a management layer, to provide services, the cloud storage system providing redundancy;

determine an at risk probability in the compute domain and a first risk index for the compute domain from the at risk probability in the compute domain;

determine an at risk probability for storage volumes in the storage domain and a second risk index for the storage domain from the at risk probability in the storage domain;

determine an at risk probability for storage paths to the storage domain from the compute domain and a third risk index for the storage paths from the at risk probability of the storage paths;

determine an overall at risk probability for a first one of the services using system resources and an overall risk index for the first one of the services from the overall risk probability; and

determine a revised overall risk index after at least two component failures in at least two of the compute domain, storage domain, and storage paths to assist a user in selecting a first one of the at least two component failures to fix first.

7. The article according to claim 6, further including instructions for using a logarithm to generate the overall risk index.

8. The article according to claim 6, wherein the at risk probability for the compute domain is derived from a number of blades used by the first one of the services, a failover rate relating to a number of servers that can fail without putting a compute cluster service into an overcommitted state, an annual failure rate for the blades used by the first one of the services, and an amount of time to replace one of the blades used by the first one of the services.

9. The article according to claim 6, wherein the at risk probability for the storage volumes is derived from a number of storage volumes used by the first one of the services, a number of disks in the storage volumes, a number of disks that can fail in the storage volumes without data loss, an annual disk failure rate, and an amount of time needed to convert a hot spare disk to a contributing disk of the storage volume.

10. The article according to claim 6, wherein the at risk probability for the storage paths is derived from a number of storage volumes used by the first one of the services, a number of blades used by the first one of the services, a number of cables used by the storage volumes, an annual failure rate for the cables, and an amount of time to replace a failed one of the cables.

11. A pre-integrated cloud computing system having a compute domain, a network domain, a storage domain, a virtual layer and a management layer, to provide services, the cloud storage system providing redundancy, comprising:

a processor; and

a memory containing instructions to enable the processor to;

determine an at risk probability in the compute domain and a first risk index for the compute domain from the at risk probability in the compute domain;

determine an at risk probability for storage volumes in the storage domain and a second risk index for the storage domain from the at risk probability in the storage domain;

determine an at risk probability for storage paths to the storage domain from the compute domain and a third risk index for the storage paths from the at risk probability of the storage paths;

determine an overall at risk probability for a first one of the services using system resources and an overall risk index for the first one of the services from the overall risk probability; and

determine a revised overall risk index after at least two component failures in at least two of the compute domain, storage domain, and storage paths to assist a user in selecting a first one of the at least two component failures to fix first.

12. The system according to claim 11, wherein the processor and memory are further configured to use a logarithm to generate the overall risk index.

13. The system according to claim 11, wherein the at risk probability for the compute domain is derived from a number of blades used by the first one of the services, a failover rate relating to a number of servers that can fail without putting a compute cluster service into an overcommitted state, an annual failure rate for the blades used by the first one of the services, and an amount of time to replace one of the blades used by the first one of the services.

14. The system according to claim 11, wherein the at risk probability for the storage volumes is derived from a number of storage volumes used by the first one of the services, a number of disks in the storage volumes, a number of disks that can fail in the storage volumes without data loss, an annual disk failure rate, and an amount of time needed to convert a hot spare disk to a contributing disk of the storage volume.

15. The system according to claim 11, wherein the at risk probability for the storage paths is derived from a number of storage volumes used by the first one of the services, a number of blades used by the first one of the services, a number of cables used by the storage volumes, an annual failure rate for the cables, and an amount of time to replace a failed one of the cables.