Automated selection of functions to reduce storage capacity based on performance requirements

Patent metadata

Field	Value
Publication number	US-10444996-B2
Application number	US-201414571187-A
Country	US
Kind code	B2
Filing date	Dec 15, 2014
Priority date	Sep 28, 2011
Publication date	Oct 15, 2019
Grant date	Oct 15, 2019

Abstract

A plurality of functions to configure a unit of a storage volume is maintained, wherein each of the plurality of functions, in response to being applied to the unit of the storage volume, configures the unit of the storage volume differently. Statistics are computed on growth rate of data and access characteristics of the data stored in the unit of the storage volume. A determination is made as to which of the plurality of functions to apply to the unit of the storage volume, based on the computed statistics.

First claim

What is claimed is:

1. A computer implemented method for automated selection of functions to reduce storage capacity of a computer storage device, comprising:
maintaining a plurality of functions to configure a unit of a storage volume, wherein each of the plurality of functions, in response to being applied to the unit of the storage volume, configures the unit of the storage volume differently, wherein the plurality of functions includes full allocation, thin provisioning, data compression, and data deduplication;
computing statistics on growth rate of data and access characteristics of the data stored in the unit of the storage volume;
determining which of the plurality of functions to apply to the unit of the storage volume, based on the computed statistics wherein the plurality of functions includes full allocation, thin provisioning, data compression, and data deduplication in a decreasing order of performance in terms of performing operations on the unit of the storage volume, and in an increasing order of efficiency in terms of storage requirements in the unit of the storage volume, wherein: a new unit is initially provisioned as fully allocated and does not use thin provisioning, data compression or data deduplication; and based on the computed statistics, migrating the new unit to use thin provisioning, data compression, or data deduplication;
applying full allocation to the unit of the storage volume, in response to determining that the access time requirement for the unit of the storage volume is less than a predetermined threshold amount of time, wherein the computed statistics include determining an access time requirement for the unit of the storage volume; and
applying thin provisioning to the unit of the storage volume, in response to determining that the rate at which write operations are performed on the unit of the storage volume does not exceed a predetermined rate of write operations and that the available amount of unused space on the unit of the storage volume exceeds a predetermined threshold, wherein the computed statistics include a rate at which write operations are performed on the unit of the storage volume and an available amount of unused space on the unit of the storage volume, wherein not too many write operations that grow the volume are taking place.

2. The method of claim 1 , wherein the computed statistics include determining how much duplicative data is present on the unit the storage volume, the method further comprising:
applying data deduplication to the unit of the storage volume, in response to determining that duplicative data present on the unit of the storage volume exceeds a predetermined threshold and access time requirement for the unit of the storage volume is greater than a predetermined access time rate.

3. The method of claim 1 , wherein the computed statistics include determining how much the data stored on the unit of the storage volume is capable of being compressed, the method further comprising:
compressing the data stored on the unit of the storage volume, in response to determining that access time requirement for the data will be met subsequent to the compressing of the data stored in the unit of the storage volume.

4. The method of claim 1 , wherein:
a first unit that is a most frequently accessed unit of the storage volume is maintained as a fully allocated unit;
a second unit that is a less frequently accessed unit of the storage volume compared to the first unit is maintained as a thin provisioned unit;
a third unit that is a still less frequently accessed unit of the storage volume compared to the second unit is maintained as a compressed unit; and
a fourth unit that is a least frequently accessed unit of the storage volume is maintained as a deduplicated unit.

5. The method of claim 1 , wherein if compression throughput rates are able to satisfy access patterns on the new unit, migrating the new unit that is initially provisioned as fully allocated to a compressed unit.

6. A system for automated selection of functions to reduce storage capacity of a computer storage device, comprising:
a memory; and
a processor coupled to the memory, wherein the processor performs operations, the operations comprising;
maintaining a plurality of functions to configure a unit of a storage volume, wherein each of the plurality of functions, in response to being applied to the unit of the storage volume, configures the unit of the storage volume differently, wherein the plurality of functions includes full allocation, thin provisioning, data compression, and data deduplication,
computing statistics on growth rate of data and access characteristics of the data stored in the unit of the storage volume,
determining which of the plurality of functions to apply to the unit of the storage volume, based on the computed statistics wherein the plurality of functions includes full allocation, thin provisioning, data compression, and data deduplication in a decreasing order of performance in terms of performing operations on the unit of the storage volume, and in an increasing order of efficiency in terms of storage requirements in the unit of the storage volume, wherein: a new unit is initially provisioned as fully allocated and does not use thin provisioning, data compression or data deduplication; and based on the computed statistics, migrating the new unit to use thin provisioning, data compression, or data deduplication,
applying full allocation to the unit of the storage volume, in response to determining that the access time requirement for the unit of the storage volume is less than a predetermined threshold amount of time, wherein the computed statistics include determining an access time requirement for the unit of the storage volume, and
applying thin provisioning to the unit of the storage volume, in response to determining that the rate at which write operations are performed on the unit of the storage volume does not exceed a predetermined rate of write operations and that the available amount of unused space on the unit of the storage volume exceeds a predetermined threshold, wherein the computed statistics include a rate at which write operations are performed on the unit of the storage volume and an available amount of unused space on the unit of the storage volume, wherein not too many write operations that grow the volume are taking place.

7. The system of claim 6 , wherein the computed statistics include determining how much duplicative data is present on the unit the storage volume, the operations further comprising:
applying data deduplication to the unit of the storage volume, in response to determining that duplicative data present on the unit of the storage volume exceeds a predetermined threshold and access time requirement for the unit of the storage volume is greater than a predetermined access time rate.

8. The system of claim 6 , wherein the computed statistics include determining how much the data stored on the unit of the storage volume is capable of being compressed, the operations further comprising:
compressing the data stored on the unit of the storage volume, in response to determining that access time requirement for the data will be met subsequent to the compressing of the data stored in the unit of the storage volume.

9. The system of claim 6 , wherein:
a first unit that is a most frequently accessed unit of the storage volume is maintained as a fully allocated unit;
a second unit that is a less frequently accessed unit of the storage volume compared to the first unit is maintained as a thin provisioned unit;
a third unit that is a still less frequently accessed unit of the storage volume compared to the second unit is maintained as a compress

Assignees

• IBM

Inventors

• Cashman Paul N
• Whyte Barry D

Classifications

• G06F3/0608 Primary

  Saving storage space on storage systems · CPC title

• G06F3/0641

  De-duplication techniques · CPC title

• G06F3/067

  Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS] · CPC title

• G06F3/0665

  at area level, e.g. provisioning of virtual or logical volumes · CPC title

• G06F3/061

  Improving I/O performance · CPC title