Method for policy-based, autonomically allocated storage

• US 7,480,912 B2
• Filed: 05/29/2003
• Issued: 01/20/2009
• Est. Priority Date: 05/29/2003
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Whalley, Ian N., Devarakonda, Murthy V., Arnold, William C., Segal, Alla, Chess, David M.
• Primary Examiner(s): An; Meng-Ai
• Assistant Examiner(s): WAI, ERIC CHARLES
• Application Number:
  US10/449,269
• Publication Number:
  US 20040243692A1
• Time in Patent Office: 2,063 Days
• Field of Search: 709/217, 718104-105, 370/320, 711/111, 714/47
• US Class Current: 718/104
• CPC Class Codes: G06F 9/5016 the resource being the memory
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/1008 based on parameters of servers, e.g. available memory or workload monitoring of computer activity G06F11/30
• CPC Class Codes: H04L 67/101 based on network conditions
• CPC Class Codes: H04L 67/1097 for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
• CPC Class Codes: H04L 67/535 Tracking the activity of the user network monitoring arrangements H04L43/00; recording of computer activity G06F11/34
• CPC Class Codes: H04L 69/329 in the application layer [OSI layer 7]

Abstract

Methods for allocation of storage resources, performance monitoring, and reallocation of resources to eliminate hot spots, by specifying high-level goals, rather than by means of low-level manual steps. Policies are specified as administrator specified constraints under which the resources are managed. Goals are specified in terms of performance, availability, and security requirements of the desired storage. As a part of the automation, this invention provides a method for analyzing capabilities of the computer storage system and forming analysis results, which are later used for determining an allocation of resources that will meet the high-level goals specified. This invention also provides methods for automatic monitoring of performance, availability, and security goals for allocated resources. If goals are not met, resources are reallocated so that the goals can be met with the allocation. The invention reduces human involvement, allows policy control, minimizes error, and provides efficient service delivery specified by policies.

Claims

1. A method for allocating storage in a computer storage system, the method comprising:

specifying constraints on resource allocation in said computer storage system including specifying a set of attributes for the quality-of-service;

analyzing available capabilities of the computer storage system and generating analysis results from said analyzing that includes building a logical model of a storage infrastructure based on configuration information collected via sensors at the time of system initialization;

associating an allocation request with a given quality-of service;

allocating a portion of available storage in the computer storage system based on the allocation request, at least one policy in said constraints, the analysis results, the quality-of-service associated with the allocation request and the logical model;

monitoring the computer storage system to determine whether the allocating of the portion of available storage in response to allocation request fails to meet the quality-of-service; and

reallocating resources in response to the determination that the portion of the available storage fails to meet the quality-of-service associated with the allocation request, wherein said reallocating includes reading and processing data derived from dynamic sensors.

2. A method as recited in claim 1, wherein the step of specifying constraints on resource allocation includes specifying at least one constraint from a group of constraints consisting of:

a quality-of-service selection for users;

a quality-of-service for applications, quality-of-service patterns, when and how resources may be allocated.

3. A method as recited in claim 1, wherein the attributes of the set of attributes are selected from a group of attributes consisting of:

maximum size;

throughput;

required response time;

expected ratio of sequential and random accesses;

read-write ratios;

availability requirements;

level of redundancy;

security requirements; and

any combination of these.

4. A method as recited in claim 1, wherein the step of analyzing capabilities of the computer storage system and generating analysis results includes selecting a capability from a group of capabilities consisting of:

capacity;

speed;

throughput, latency, seek time, and any combination of these.

5. A method as recited in claim 1, wherein the logical model includes information selected from groups of information consisting of:

storage space availability;

configuration format;

number of physical paths existing from a storage subsystem to a host;

logical volumes;

mapping of physical storage; and

any combination of this information; and

wherein the capabilities of the storage system include at least one capability selected from a group of storage system levels of performance;

storage system security;

present commitments in terms of quality-of-service criteria; and

system configuration methodology to deliver a particular level of quality-of-service; and

any combination of these capabilities.

6. A method as recited in claim 5, wherein the step of analyzing capabilities includes determining whether three storage RAID arrays exist respectively having 200 MB/s, 180/MB/s, and 20 MB/d throughput availability, and whether the third array has been completely committed to serving previous allocations, and whether the first and second arrays are available on an IBM FastT storage subsystem via a McData switch.

7. A method as recited in claim 1, wherein the step of associating an allocation request with a given quality-of service includes:

selecting the quality-of-service level from at least one of;

a previously defined quality-of service levels;

storage based on observed usage of allocated storage;

management software based on business level policies and service level agreements;

any combination of these.

8. A method as recited in claim 1, wherein the step of allocating a portion of available storage in the computer includes selecting a best throughput fit way of utilizing at least part of the computer storage system to meet the allocation request based on said analysis results.

9. A method as recited in claim 1, wherein the step of monitoring the computer storage system includes observing various attributes of the computer storage system.

10. A method as recited in claim 1, wherein the step of monitoring the computer storage system to determine whether the allocating fails to meet the quality-of-service includes the steps of:

examining results generated by the steps of monitoring; and

comparing the examined monitoring results to a particular quality of service associated with each allocated portion of said available storage.

11. A method as recited in claim 1, wherein the step of reallocating resources includes the steps of:

determining a new location on the computer storage system to relocate a piece of storage currently that failed to meet the quality-of-service; and

storing the piece of storage in the new location on the computer storage system.

12. A method as in claim 1, wherein the steps of associating and allocating employ a quality-of service specification in a policy specification comprising performance specifications.

13. A method as in claim 12, wherein the performance specifications includes a specification taken from a group of specifications consisting of:

throughput, response time, transfer rate, and any combination of these.

14. A method as in claim 1, wherein the steps of associating and allocating employ a quality-of-service specification in a policy specification comprising availability specifications.

15. A method as in claim 14, wherein the availability specifications includes a specification taken from a group of specifications consisting of:

storage system up time and error rate.

16. A method as in claim 1, wherein the steps of associating and allocating employ a quality-of-service specification in a policy specification comprising space specifications.

17. A method as in claim 16, wherein the space specifications includes a specification taken from a group of specifications consisting of:

maximum size, initial size, size increments, guaranteed size and any combination of these.

18. A method as in claim 1, wherein the step of allocating employs a policy specification comprising specifying when and how allocated resources are reallocated in the step of reallocating.

19. A method as in claim 2, further comprising creating at least one standard named quality-of service specifications.

20. A method as in claim 19, further comprising employing said at least one standard named quality-of-service specifications to specify the desired quality-of-service in the step of allocating.

21. A method for allocating storage in a computer storage system, the method comprising:

a policy specification step, wherein constraints on resource allocation are specified;

the constraints being in a form of quality-of-service selection for certain users, applications, and patterns, and/or in a form of constraints upon when and how resources may be allocated including specifying a set of attributes for the quality-of-service;

a resource-analysis step, wherein available capabilities of the storage system are analyzed, and results of an analysis are saved that includes building a logical model of a storage infrastructure based on configuration information collected via sensors at the time of system initialization;

a quality-of-service specification step, wherein an allocation request is associated with a given quality-of-service;

an allocation step, wherein at least some available storage in the computer storage system is allocated based on the allocation request, on policies specified in the policy specification step, responsive to the analysis results saved in the resource-analysis step including the logical model, the quality-of-service specified in the quality-of-service specification step; and

a reallocation step, wherein resources may be reallocated in response to monitoring of the computer system that shows failure of meeting the given quality-of-service, including reading and processing data derived from dynamic sensors.