US 10,095,418 B1Granted

Automatic tiering of storage using dynamic grouping

AssigneeInternational Business Machines Corporation

InventorsYongjie Gong, Shuang Shuang Li, Yang Liu, Mei Mei, Xue Qiang Zhou

Life of this patent

Estimating the term…

Loading patent…

Abstract

In a hierarchical storage system, blocks of data selected for auto-tiering migration, are selected based on dynamically adjusted group sizes. Contiguous blocks are organized into default groups. I/O activity of the blocks in a group is monitored. Based on the I/O activity, the default groups may be sub-divided into smaller sub-groups or combined into larger groups, to separate as much as practical, contiguous series of cooler blocks and contiguous series of hotter blocks into respective focused (concentrated) groups or sub-groups. The concentrated group or sub-group may then be migrated according to the average I/O activity of the included blocks. Group configurations are continually and dynamically adjusted according to changing I/O conditions.

A method comprising:

receiving a first input/output (I/O) statistics dataset with respect to a multi-tier storage system, where the multi-tier storage system comprises:
(i) a first storage tier and a second storage tier, (ii) a plurality of data groups including a first data group, wherein the first data group comprises a first data block and the first data block is stored on the first storage tier, wherein (a) the first I/O statistics dataset includes information pertaining to a first activity metric value corresponding to I/O activity with respect to the first data block and (b) the first I/O statistics dataset includes information collected over a first time interval;
determining a first group score with respect to the first data group, where the first group score is based, at least in part, on the first activity metric value;
determining that the first group score is above an upper threshold value;
in response to determining that the first group score is above the upper threshold value, partitioning the first data group into a plurality of data sub-groups including a first data sub-group, where the first data sub-group includes the first data block;
migrating the first data sub-group to the second storage tier;
receiving a second I/O statistics dataset with respect to the multi-tier storage system, wherein (a) the second I/O statistics dataset includes information pertaining to a second activity metric value corresponding to I/O activity with respect to the first data block and (b) the second I/O statistics dataset includes information collected over a second time interval;
determining a second group score with respect to the first data sub-group, where the second group score is based, at least in part, on the second activity metric value;
determining that the second group score does not exceed a lower threshold value;
in response to determining that the second group score does not exceed the lower threshold value, migrating the first data sub-group to the first storage tier;
and
re-combining the plurality of data sub-groups into a re-combined first data group.

9 more elements

Facts

Application

15/846,767Filed 19 Dec 2017 · continuation

Priority

27 Apr 2017

Published

US 2018/0314422 A11 Nov 2018

Granted

9 Oct 2018

Term ends

Not estimatedThe source records don’t support an estimate.

Classification7 codes

G06F 3/061
Specifically adapted to achieve a particular effect: improving I/O performance
G06F 3/0619
Improving the reliability of storage systems: in relation to data integrity, e.g. data losses, bit errors
G06F 3/0647
Horizontal data movement in storage systems, i.e. moving data in between storage devices or systems: migration mechanisms
G06F 3/0665
Virtualisation aspects: at area level, e.g. provisioning of virtual or logical volumes

Abstract

Claims1