Computer system and method of managing storage system monitoring access performance for risky pool detection

1. A computer system comprising:

a storage system; and

a management system coupled to the storage system via a network and configured to manage the storage system,

wherein the storage system includes a plurality of pools each of which includes a plurality of storage resources and is separated into a plurality of tiers different in access performance;

wherein the management system monitors access performance of the plurality of pools to detect a risky pool which is expected to become unable to satisfy predetermined access performance requirements;

wherein the management system estimates access performance of each of one or more candidate pools in the plurality of pools in a case where a storage resource thereof migrates to the risky pool, by simulation based on variations in data arrangement in the pool caused by evacuating data held in the storage resource in the same pool for the migration of the storage resource; and

wherein the management system determines that the storage resource migrates from a source pool of which the estimated access performance satisfies predetermined performance requirements to the risky pool.

2. A computer system according to claim 1 , wherein the management system estimates the access performance of each of the one or more candidate pools during a period of the data evacuation using a predicted data arrangement varied by data relocation for the storage resource to migrate and estimated additional load caused by the data relocation for the storage resource to migrate.

3. A computer system according to claim 2 ,

wherein, in the estimation of the time variation of the access performance by simulation, the management system estimates the access performance of each of the one or more candidate pools in a period in which data relocation does not occur in the candidate pool using predicted data arrangement in the candidate pool; and

wherein, in the estimation of the time variation of the access performance by simulation, the management system estimates the access performance of each of the one or more candidate pools in a period in which data relocation occurs in the candidate pool using estimated additional load caused by the data relocation in the candidate pool in addition to predicted data arrangement in the candidate pool.

4. A computer system according to claim 3 , wherein the management system repeats the estimation of the access performance of each of the one or more candidate pools while advancing the start time of the data evacuation to find a start time of data evacuation with which the estimated access performance satisfies the predetermined performance requirements of the candidate pool.

5. A computer system according to claim 4 , wherein the management system selects a candidate pool in which an expected completion time of the data evacuation is the earliest as the source pool from the one or more candidate pools in which the estimated access performance satisfies the predetermined performance requirements.

6. A computer system according to claim 5 ,

wherein the management system holds history information on accesses to each of the one or more candidate pools collected from the storage system and information on access performance of individual tiers in each of the one or more candidate pools; and

wherein the management system estimates the access performance of each of the one or more candidate pools using the access history information and the information on access performance in periods during the data evacuation and after the completion of the data evacuation.

7. A computer system according to claim 6 , wherein, in a case where the simulation includes an inter-tier rearrangement after the completion of evacuating the data in the storage resource, the management system estimates access performance of each of the one or more candidate pools during the inter-tier rearrangement after the completion of the data evacuation using predicted data arrangement changed by a tier-to-tier data relocation in the candidate pool and estimated additional load caused by the tier-to-tier data relocation.

8. A computer system according to claim 7 , wherein the management system estimates access performance of each of the one or more candidate pools in both of a case where the inter-tier rearrangement is carried out immediately after evacuating the data in the storage resource and a case where the inter-tier rearrangement is not carried out immediately after evacuating the data in the storage resource.

9. A computer system according to claim 8 ,

wherein the management system monitors accesses to the source pool while the storage resource is migrating from the source pool to the risky pool; and

wherein the management system determines whether to abort the migration of the resource based on the monitored accesses to the source pool and the estimated access performance of the source pool.

10. A method of managing a storage system in a computer system, the computer system including the storage system which includes a plurality of storage resources and a plurality of pools separated into a plurality of tiers different in access performance and a management system connected to the storage system via a network to manage the storage system, the method comprising:

monitoring, by the management system, access performance of the plurality of pools to detect a risky pool which is expected to become unable to satisfy predetermined access performance requirements;

estimating, by the management system, access performance of each of one or more candidate pools in the plurality of pools in a case where a storage resource migrates to the risky pool, by simulation based on variations in data arrangement in the candidate pool caused by evacuating data held in the storage resource in the same pool for the migration of the storage resource; and

determining, by the management system, that the storage resource migrates from a source pool of which the estimated access performance satisfies predetermined performance requirements to the risky pool.

11. A method of managing a storage system according to claim 10 , further comprising:

estimating, by the management system, the access performance of each of the one or more candidate pools during a period of the data evacuation using predicted data arrangement varied by data relocation for the storage resource to migrate and estimated additional load caused by the data relocation for the storage resource to migrate.

12. A method of managing a storage system according to claim 10 , further comprising:

repeating, by the management system, the estimating the access performance of each of the one or more candidate pools while advancing the start time of the data evacuation to find a start time of data evacuation with which the estimated access performance satisfies the predetermined performance requirements; and

selecting, by the management system, a candidate pool in which an expected completion time of the data evacuation is the earliest as the source pool from the one or more candidate pools in which the estimated access performance satisfies the predetermined performance requirements.

13. A method of managing a storage system according to claim 10 , further comprising:

storing, by the management system, history information on accesses to each of the one or more candidate pools collected from the storage system and information on access performance of individual tiers in each of the one or more candidate pools; and

estimating, by the management system, access performance of each of the one or more candidate pools using the access history information and the information on access performance in periods during the data evacuation and after the completion of the data evacuation.

14. A method of managing a storage system according to claim 10 , further comprising:

estimating, by the management system, the access performance of each of the one or more candidate pools during an inter-tier rearrangement after the completion of the data evacuation using predicted data arrangement changed by a tier-to-tier data relocation in the candidate pool and estimated additional load caused by the tier-to-tier data relocation in a case where the simulation includes the inter-tier rearrangement after the completion of evacuating the data in the storage resource.

15. A method of managing a storage system according to claim 14 , further comprising:

estimating, by the management system, access performance of each of the one or more candidate pools in both of a case where the inter-tier rearrangement is carried out immediately after evacuating the data in the storage resource and a case where the inter-tier rearrangement is not carried out immediately after evacuating the data in the storage resource.