Backup performance in storage tiers using data allocation optimization

Patent metadata

Field	Value
Publication number	US-10768848-B2
Application number	US-201514754094-A
Country	US
Kind code	B2
Filing date	Jun 29, 2015
Priority date	Jun 29, 2015
Publication date	Sep 8, 2020
Grant date	Sep 8, 2020

Abstract

A system for improving backup performance includes an interface, a processor, and a memory. The interface is to receive a volume identifier to backup. The processor is to determine allocation for data of a volume associated with the volume identifier in a plurality of storage tiers to improve backup processing, move data based at least in part on the allocation, and backup data of the volume. The memory is coupled to the processor and configured to provide the processor with instructions.

First claim

What is claimed is:

1. A system for improving backup performance comprising:
an interface to:
receive a volume identifier to backup;

a processor to:
determine a first allocation for backup for data of a volume associated with the volume identifier in a plurality of storage tiers, comprising:
determining a plurality of storage tiers with capacity to store the data,
for each storage tier of the plurality of storage tiers with capacity to store the data other than a current storage tier:
determining a benefit in backup time based on a speed difference between the current storage tier and each storage tier and a maximum possible amount of data that can be moved within available time, the maximum possible amount of data that can be moved being determined based on an amount of time remaining until a next backup and a bandwidth available to move data between tiers, and
determining a cost based on time lost due to moving the data between the current storage tier and each storage tier using a cost function;

determining the first allocation, wherein the first allocation is associated with a highest benefit in backup time to cost ratio;

move the data based on the determined first allocation to improve backup processing;
in response to moving the data based on the determined first allocation, back up the data of the volume;
modify the cost function based on a success of the first allocation for data of the volume for determining a subsequent first allocation;
determine a second allocation for normal operational access different from the first allocation for the data of the volume;
subsequent to the moving of the data based on the determined first allocation and the backing up, move the data based on the determined second allocation to optimize access to the data; and

a memory coupled to the processor and configured to provide the processor with instructions.

2. The system as in claim 1 , wherein determining the first allocation comprises determining available storage in a plurality of storage tiers.

3. The system as in claim 1 , wherein determining the first allocation is based at least in part on percentage of data that has changed in the volume.

4. The system as in claim 1 , wherein determining the first allocation is based at least in part on a number of storage tiers.

5. The system as in claim 1 , wherein determining the first allocation is based at least in part on performance of each of the plurality of storage tiers.

6. The system as in claim 1 , wherein determining the first allocation comprises determining the cost function.

7. The system as in claim 6 , wherein the cost function includes a cost of moving data between a first tier and a second tier.

8. The system as in claim 7 , wherein the cost of moving data is based at least in part on a moving time.

9. The system as in claim 7 , wherein the cost of moving data is based at least in part on a system load.

10. The system as in claim 7 , wherein the cost of moving is based at least in part on a bandwidth limit for moving.

11. The system as in claim 6 , wherein the cost function is based at least in part on the current storage tier of the data.

12. The system as in claim 1 , wherein the plurality of storage tiers includes a slower storage tier and a faster storage tier.

13. The system as in claim 12 , further comprising moving data from the slower storage tier to the faster storage tier.

14. The system as in claim 12 , further comprising moving data from the faster storage tier to the slower storage tier.

15. The system as in claim 1 wherein the processor modifies the cost function by modifying one or more weights of one or more weight factors of the cost function, wherein the weights are adjusted over time based on information of one or more backups, and wherein the one or more weight factors comprise one or more of: a percentage of data changed since last backup, a time to next backup, a bandwidth between storage tiers, an available space in each tier, a time for transferring data between the tiers, or savings in backup time from moving the data to faster tiers.

16. A method of improving backup performance, comprising:
receiving a volume identifier to backup;
determining, using a processor, a first allocation for backup for data of a volume associated with the volume identifier in a plurality of storage tiers, comprising:
determining at least one higher storage tier with capacity to store the data,
for each storage tier of the plurality of storage tiers with capacity to store the data other than a current storage tier:
determining a benefit in backup time based on a speed difference between the current storage tier and each storage tier and a maximum possible amount of data that can be moved within available time, the maximum possible amount of data that can be moved being determined based on an amount of time remaining until a next backup and a bandwidth available to move data between tiers, and
determining a cost based on time lost due to moving the data between the current storage tier and each storage tier using a cost function;

determining the first allocation, wherein the first allocation is associated with a highest benefit in backup time to cost ratio;

moving the data based on the determined first allocation to improve backup processing;
in response to moving the data based on the determined first allocation, backing up the data of the volume; modifying the cost function based on a success of the first allocation for data of the volume for determining a subsequent first allocation;
determining a second allocation for normal operational access different from the first allocation for the data of the volume; and
subsequent to the moving of the data based on the determined first allocation and the backing up, moving the data based on the determined second allocation to optimize access to data.

17. A computer program product for improving backup performance, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:
receiving a volume identifier to backup;
determining, using a processor, a first allocation for backup for data of a volume associated with the volume identifier in a plurality of storage tiers, comprising:
determining at least one higher storage tier with capacity to store the data,
for each storage tier of the plurality of storage tiers with capacity to store the data other than a current storage tier:
determining a benefit in backup time based on a speed difference between the current storage tier and each storage tier and a maximum possible amount of data that can be moved within available time, the maximum possible amount of data that can be moved being determined based on an amount of time remaining until a next backup and a bandwidth available to move data between tiers, and
determining a cost based on time lost due to moving the data between the current storage tier and each storage tier using a cost function;

determining the first allocation, wherein the first allocation is associated with a highest benefit in backup time to cost ratio;

moving the data based on the first allocation to improve backup processing;
in response to moving the data based on the determined first allocation, backing up the data of the volume;
modifying the cost function based on a success of the first allocation for data of the volume for determining a subsequent first allocation;
determining a second allocation for normal operational access different from the first allocation for the data of the vol

Assignees

• Emc Corp
• Emc Ip Holding Co Llc

Inventors

• Wigmore Ian
• Smaldone Stephen D
• Marshak Marik
• Don Arieh

Classifications

• G06F3/061 Primary

  Improving I/O performance · CPC title

• G06F3/0685

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

• G06F11/1461

  Backup scheduling policy · CPC title

• G06F3/065 Primary

  Replication mechanisms · CPC title

• G06F11/1448 Primary

  Management of the data involved in backup or backup restore · CPC title