Identifying workload and sizing of buffers for the purpose of volume replication

Patent metadata

Field	Value
Publication number	US-9854037-B2
Application number	US-201314770605-A
Country	US
Kind code	B2
Filing date	May 13, 2013
Priority date	May 13, 2013
Publication date	Dec 26, 2017
Grant date	Dec 26, 2017

Abstract

A controller is operable to: identify virtual machines to be protected in a first storage system; identify logical volumes used by the virtual machines based on first relationship information; calculate workload, based on information of workload monitored for the identified logical volumes; and calculate size of a buffer area in the first storage system to be used for temporarily storing copy data to be sent to a second storage system in remote copy procedure of one or more remote copy pairs, based on the calculated workload, each copy pair being formed by a logical volume of the identified logical volumes in the first storage system as primary logical volume and another logical volume in the second storage system as secondary logical volume, so that the buffer area having a size equal to or greater than the calculated size can be used to manage protection of the identified virtual machines.

First claim

What is claimed is:

1. A computer comprising:
a controller; and
a memory connected to the controller and which stores instructions that, when executed by the controller, cause the controller to:
identify one or more virtual machines of a plurality of virtual machines to be protected in a first storage system;
identify one or more logical volumes of a plurality of logical volumes in the first storage system, the one or more logical volumes used by the
identified one or more virtual machines, based on first relationship information between the plurality of virtual machines and the plurality of logical volumes, the first relationship information being managed by the controller;
calculate a workload to be used for the identified one or more logical volumes, based on information of workload monitored for each of the identified one or more logical volumes, the information of workload monitored being managed by the controller; and
calculate a size of a buffer area in the first storage system to be used for temporarily storing copy data to be sent from the first storage system to a secondstorage system in a remote copy procedure of one or more remote copy pairs, based on the calculated workload of the identified one or more logical volumes, each of the one or more remote copy pairs being formed by a logical volume of the identified one or more logical volumes in the first storage system as a primary logical volume and another logical volume in the second storage system as a secondary logical volume, so that the buffer area for the remote copy procedure of the identified one or more logical volumes having a size equal to or greater than the calculated size can be used to manage protection of the identified one or more virtual machines.

2. The computer according to claim 1 ,
wherein the controller is further caused to (i) create a new buffer area in the first storage system based on the calculated workload and the calculated size of the buffer area or (ii) modify an existing buffer area in the first storage system based on the calculated workload and the calculated size of the buffer area.

3. The computer according to claim 1 ,
wherein the controller is further caused to (i) create a new buffer area in the second storage system based on the calculated workload and the calculated size of the buffer area or (ii) modify an existing buffer area in the second storage system based on the calculated workload and the calculated size of the buffer area.

4. The computer according to claim 1 ,
wherein the first storage system has an overhead and an outage duration; and
wherein the size of the buffer area is calculated as the calculated workload multiplied by the outage duration multiplied by (1+overhead).

5. The computer according to claim 1 ,
wherein the first storage system has a plurality of disk groups including used disk groups and unused disk groups;
wherein each unused disk group has a disk group layout with a disk group size and a disk group throughput;
wherein the controller is further caused to select, from the unused disk groups, one or more disk groups to be used by the buffer area in the first storage system, and the selection of the one or more disk groups comprises:
organizing the unused disk groups by disk group layout;
identifying, for each disk group layout, a list of candidate disk groups which have a total disk group throughput that is greater than the calculated workload and which have a total disk group size that is greater than the calculated size of the buffer area, so as to produce one or more lists of candidate disk groups;
selecting, from the identified one or more lists of candidate disk groups, the list of candidate disk groups having the highest total disk group throughput; and
determining a number of disk groups required, in the selected list of candidate disk groups, for the buffer area by (i) dividing the calculated workload by the total disk group throughput for the selected list and rounding up to the nearest whole number to obtain a first calculated value, (ii) dividing the calculated size of the buffer area by the disk group size of the selected list and rounding up to the nearest whole number to obtain a second calculated value, and (iii) selecting a maximum of the first calculated value and the second calculated value as the number of disk groups required in the selected list of candidate disk groups for the buffer area.

6. The computer according to claim 1 ,
wherein the calculated size of the buffer area is for modifying an existing buffer area which has a list of existing one or more virtual machines to protect; and
wherein the controller is further caused to:
identify, from the list of existing one or more virtual machines, which of the existing one or more virtual machines will retain protection; and
identify any one or more virtual machines to be newly protected;
wherein the one or more logical volumes are identified which are used by the identified one or more virtual machines to retain protection and the identified one or more virtual machines to be newly protected.

7. The computer according to claim 1 ,
wherein the first storage system has a plurality of disk groups; and
wherein the calculated size of the buffer area is for selecting an existing buffer area which has a size that is greater than the calculated size of the buffer area and which has a total throughput that is greater than the calculated workload, the total throughput of the existing buffer area being equal to a product of a disk group throughput of one or more disk groups allocated to the existing buffer area and a number of the one or more disk groups allocated to the existing buffer area.

8. The computer according to claim 1 ,
wherein the first storage system has a plurality of disk groups including used disk groups and unused disk groups;
wherein the calculated size of the buffer area is for modifying an existing buffer area (i) which has a size that is not greater than the calculated size of thebuffer area or (ii) which has a total throughput that is not greater than the calculated workload, the total throughput of the existing buffer area being equal to a product of a disk group throughput of one or more existing disk groups allocated to the existing buffer area and a number of the one or more existing disk groups allocated to the existing buffer area; and
wherein the controller is further caused to select, from the unused disk groups, one or more disk groups to be added to the existing buffer area in the first storage system.

9. The computer according to claim 8 ,
wherein each unused disk group has a disk group layout with a disk group size and a disk group throughput;
wherein the one or more existing disk groups in the existing buffer area have an existing disk group layout; and
wherein the selection of the one or more disk groups to be added to the existing buffer area in the first storage system includes:
organizing the unused disk groups by disk group layout;
identifying, from the unused disk groups which have a same disk group layout as the existing disk group layout, a list of candidate disk groups which, when added to the existing disk groups, have a total disk group throughput that is greater than thecalculated workload and which have a total disk group size that is greater than the calculated size of the buffer area; and
determining a number of disk groups required, in the identified list of candidate disk groups, to be added to the existing buffer area by (i) dividing the calculated workload by the total disk group throughput for the identified list when added to the existing disk groups and rounding up to the nearest whole number to obtain a first calculated value, (ii) dividing the calcula

Assignees

• Hitachi Ltd

Inventors

• Murrish Randall
• Walker Steven
• Luraghi Oswald
• Liu Christopher

Classifications

• G06F11/3433

  for load management (allocation of a server based on load conditions G06F9/505; load rebalancing G06F9/5083; redistributing the load in a network by a load balancer H04L67/1029) · CPC title

• G06F11/3442

  for planning or managing the needed capacity · CPC title

• G06F11/2071

  using a plurality of controllers · CPC title

• G06F3/0617 Primary

  in relation to availability · CPC title

• G06F3/065

  Replication mechanisms · CPC title