Compression of host I/O data in a storage processor of a data storage system with selection of data compression components based on a current fullness level of a persistent cache

Patent metadata

Field	Value
Publication number	US-11068405-B2
Application number	US-201815957065-A
Country	US
Kind code	B2
Filing date	Apr 19, 2018
Priority date	Apr 19, 2018
Publication date	Jul 20, 2021
Grant date	Jul 20, 2021

Abstract

A storage processor in a data storage system includes a compression selection component that selects a data compression component to be used to compress host I/O data that is flushed from a persistent cache of the storage processor based on a current fullness level of the persistent cache. The compression selection component selects compression components implementing compression algorithms having relatively lower compression ratios for relatively higher current fullness levels of the persistent cache, and selects compression components implementing compression algorithms having relatively higher compression ratios for relatively lower current fullness levels of the persistent cache.

First claim

What is claimed is:

1. A method of providing data compression in a storage processor of a data storage system, comprising the steps of:
in response to detecting a cache flush event by detecting that a predetermined time period has expired since host I/O data was previously stored into a persistent cache located in the storage processor, i) forming an aggregation set of blocks of host I/O data within host I/O data accumulated in the persistent cache, wherein the aggregation set is a set of oldest blocks of host I/O data that are stored in the persistent cache, and ii) determining a current fullness level of the persistent cache, wherein the current fullness of the persistent cache comprises a current percentage of a total size of the persistent cache that is currently used to store host I/O data;
selecting, by a compression selection component in the storage processor in response to the current fullness level of the persistent cache, from a set of available compression components contained in the storage processor, a compression component for compressing the aggregation set, wherein the compression selection component selects compression components implementing compression algorithms having relatively lower compression ratios in response to relatively higher current fullness levels of the persistent cache, and wherein the compression selection component selects compression components implementing compression algorithms having relatively higher compression ratios in response to relatively lower current fullness levels of the persistent cache; and
compressing the aggregation set using the selected compression component to obtain a compressed version of the aggregation set.

2. The method of claim 1 , further comprising:
accumulating host I/O data that is indicated by host I/O write requests that are received by the storage processor from at least one host computer into the persistent cache that is located in the storage processor; and
wherein the cache flush event indicates that a plurality of blocks of host I/O data accumulated in the persistent cache are to be flushed from the persistent cache to at least one lower deck file system that is internal to the data storage system.

3. The method of claim 2 , wherein selecting the compression component comprises selecting one of the compression components in the set of available compression components having the highest associated persistent cache fullness threshold that is less than or equal to the current fullness level of the persistent cache.

4. The method of claim 3 , wherein determining the current fullness level of the persistent cache comprises determining a current percentage of the total size of the persistent cache that is currently used to store host I/O data; and
wherein selecting the compression component comprises selecting one of the compression components in the set of available compression components in the set of available compression components having the highest associated persistent cache fullness threshold that is less than or equal to the current percentage of the total size of the persistent cache that is currently used to store host I/O data.

5. The method of claim 4 , wherein each one of the compression components in the set of available compression components is associated with a different persistent cache fullness threshold.

6. The method of claim 5 , further comprising:
determining whether the blocks of host I/O data in the aggregation set contain host I/O data that is an overwrite of host I/O data previously flushed from the persistent cache to at least one lower deck file system that is internal to the data storage system;
in response to determining that the blocks of host I/O data in the aggregation set contain host I/O data that is not an overwrite of host I/O data previously flushed from the persistent cache to at least one lower deck file system that is internal to the data storage system, allocating an extent of blocks in an address space of a lower deck file system within a compressed region of the address space of the lower deck file system; and
storing the compressed version of the aggregation set to non-volatile storage corresponding to the extent of blocks in the address space of the lower deck file system.

7. The method of claim 6 , further comprising:
determining a length of the compressed version of the aggregation set; and
wherein allocating the extent of blocks in the address space of the lower deck file system within the compressed region of the address space of the lower deck file system comprises allocating an extent of blocks in the address space of the lower deck file system having length that is greater than the length of the compressed version of the aggregation set.

8. The method of claim 5 , further comprising:
determining whether the blocks of host I/O data in the aggregation set contain host I/O data that is an overwrite of host I/O data previously flushed from the persistent cache to at least one lower deck file system that is internal to the data storage system;
in response to determining that the blocks of host I/O data in the aggregation set contain host I/O data that is an overwrite of host I/O data previously flushed from the persistent cache to at least one lower deck file system that is internal to the data storage system, comparing a length the compressed version of the aggregation set to a length of an extent of blocks in an address space of a lower deck file system within a compressed region of the address space of the lower deck file system that was used to store a previous compressed version of the blocks of host I/O data in the aggregation set; and
in response to determining that the length the compressed version of the aggregation set is greater than the length of an extent of blocks in an address space of the lower deck file system within the compressed region of the address space of the lower deck file system that was used to store the previous compressed version of the blocks of host I/O data in the aggregation set, selecting a second compression component for compressing the aggregation set, wherein the second selected compression component has a higher compression ratio than the initially selected compression component; and
recompressing the aggregation set using the second selected compression component to obtain a second compressed version of the aggregation set.

9. The method of claim 5 , wherein the lower deck file system maps files to blocks in an address space of the lower deck file system, wherein each file implements a corresponding data object that is accessible to the at least one host computer, and wherein blocks in the address space of the lower deck file system also correspond to non-volatile storage provisioned from a set of storage devices in the data storage system that provide storage for the lower deck file system.

10. A storage processor in a data storage system, the storage processor including a set of processing units and memory, coupled to the set of processing units, the memory storing instructions for providing data compression in the storage processor, wherein the instructions, when executed by the set of processing units, cause the set of processing units to:
in response to detection of a cache flush event by detecting that a predetermined time period has expired since host I/O data was previously stored into a persistent cache located in the storage processor, i) form an aggregation set of blocks of host I/O data within host I/O data accumulated in the persistent cache, wherein the aggregation set is a set of oldest blocks of host I/O data that are stored in the persistent cache, and ii) determine a current fullness level of the persistent cache, wherein the current ful

Assignees

• Emc Ip Holding Co Llc

Inventors

• Armangau Philippe
• Bassov Ivan
• Chaudhary Monica
• Seibel Christopher A

Classifications

• G06F12/0891 Primary

  using clearing, invalidating or resetting means · CPC title

• G06F2212/502

  using adaptive policy · CPC title

• G06F3/0673

  Single storage device · CPC title

• G06F12/1408

  by using cryptography (for digital transmission H04L9/00) · CPC title

• G06F2212/222

  Non-volatile memory · CPC title