Red Hat Summit Red Hat Summit지원콘솔개발자평가판 시작연락처

이 콘텐츠는 선택한 언어로 제공되지 않습니다.

Chapter 5. BlueStore server side compression

BlueStore server-side compression reduces raw disk usage by compressing data at the storage layer based on pool configuration and object characteristics.

BlueStore applies compression based on pool settings, the configured compression mode, and object access patterns. Each blob (data fragment) is evaluated individually to determine whether compression is applied.

Object characteristics determine which blob size BlueStore uses, not whether compression is enabled.

  • For immutable, append-only objects, or objects with sequential read access, BlueStore uses the compression_max_blob_size parameter.
  • For objects with random read or write access, BlueStore uses the compression_min_blob_size parameter.

Benefits

BlueStore server-side compression provides the following benefits:

  • Stores more data without adding physical capacity.
  • Reduces raw disk usage and infrastructure costs.
  • Enables selective compression per pool to balance performance and efficiency.

Considerations

Compression introduces CPU overhead and can slightly increase read and write latency. In some scenarios, small overwrites on compressed objects can increase space usage because overlapping data fragments require both the old and new blobs to be retained.

Enable compression only for pools or workloads where data reduction benefits outweigh performance costs. Avoid enabling compression cluster-wide without testing workload access patterns.

How compression works

BlueStore applies compression based on pool settings and object characteristics. Each blob is evaluated independently.

Compression is applied when an object meets the following conditions:

  • Marked as immutable or append-only
  • Not expected to have random reads or writes
  • Expected to have sequential reads

If these conditions are met, BlueStore uses the compression_max_blob_size parameter. Otherwise, it uses the compression_min_blob_size parameter.

Blob size determines how data is divided and aligned before compression. Larger blobs typically improve compression efficiency, while smaller blobs improve read performance by reducing the amount of data that must be decompressed.

Figure 1 illustrates how BlueStore selects a blob size based on object properties.

BlueStore compression workflow

BlueStore does not use a continuous range of blob sizes. Instead, it selects one of the two configured blob sizes based on allocation hints provided through the rados_set_alloc_hint2 API call.

Ceph clients such as RBD, CephFS, and Ceph Object Gateway (RGW) already provide these hints to optimize storage performance.

Compression behavior

BlueStore compresses write operations according to the configured compression mode, required ratio, and blob size limits.

The supported compression modes are:

  • none – Compression is never applied.
  • passive – Compression is applied only when the client marks data as compressible.
  • aggressive – Compression is applied unless the client marks data as incompressible.
  • force – Compression is always attempted.

The compression_required_ratio parameter defines the minimum acceptable compression ratio. For example, when set to 0.7, BlueStore compresses data only if the compressed output is 70 percent or smaller than the original data size. If the threshold is not met, the data is stored uncompressed.

This ratio check is performed for each blob individually to prevent unnecessary CPU usage and space amplification.

Blob size is controlled by the compression_min_blob_size and compression_max_blob_size parameters. BlueStore supports the following compression algorithms:

  • snappy
  • zlib
  • lz4
  • zstd

When configured appropriately, BlueStore balances capacity savings and performance efficiency across Red Hat Ceph Storage clusters.

5.1. Configuring BlueStore compression
링크 복사

Learn how to enable, adjust, or remove BlueStore compression settings to optimize storage efficiency and performance.

5.1.1. Configuring compression for a specific pool
링크 복사

Override global defaults by setting compression options for individual pools to fine-tune performance and data reduction.

Procedure

  1. Set the compression algorithm for the pool. 

    ceph osd pool set POOL_NAME compression_algorithm ALGORITHM
    Copy to Clipboard Toggle word wrap

    Example

    [ceph: root@host01 /]# ceph osd pool set test_pool compression_algorithm snappy
set pool 2 compression_algorithm to snappy
    Copy to Clipboard Toggle word wrap

  2. Set the compression mode for the pool. 

    ceph osd pool set POOL_NAME compression_mode MODE
    Copy to Clipboard Toggle word wrap

    Example

    [ceph: root@host01 /]# ceph osd pool set test_pool compression_mode force
set pool 2 compression_mode to force
    Copy to Clipboard Toggle word wrap

  3. Set the required compression ratio for the pool.

    Note

    Compression occurs only when compressed data size is less than or equal to 70% of the original size. 

    ceph osd pool set POOL_NAME compression_required_ratio RATIO
    Copy to Clipboard Toggle word wrap

    Example

    [ceph: root@host01 /]# ceph osd pool set test_pool compression_required_ratio 0.7
    Copy to Clipboard Toggle word wrap

  4. Set the minimum and maximum blob sizes for the pool. 

    ceph osd pool set POOL_NAME compression_min_blob_size SIZE
ceph osd pool set POOL_NAME compression_max_blob_size SIZE
    Copy to Clipboard Toggle word wrap

  5. Verify the per-pool compression settings. 

    ceph osd pool get POOL_NAME OPTION_NAME
    Copy to Clipboard Toggle word wrap

    Example

    [ceph: root@host01 /]# ceph osd pool get test_pool compression_algorithm
compression_algorithm: snappy
    Copy to Clipboard Toggle word wrap

5.1.2. Configuring global compression
링크 복사

Set cluster-wide BlueStore compression policies that apply to all pools unless overridden locally.

Procedure

  1. Set the global compression algorithm. 

    ceph config set osd bluestore_compression_algorithm ALGORITHM
    Copy to Clipboard Toggle word wrap

The default value is snappy. This parameter is optional.

  1. Set the global compression mode. 

    ceph config set osd bluestore_compression_mode MODE
    Copy to Clipboard Toggle word wrap

    The supported MODE inputs are:

    • none (default)
    • passive
    • aggressive
    • force

  2. Set the global compression ratio. 

    ceph config set osd bluestore_compression_required_ratio RATIO
    Copy to Clipboard Toggle word wrap

    The default value is 0.875. A value of 0.7 means that compression is applied only if the compressed data is 70% or smaller than the original.

  3. Set the minimum and maximum blob sizes. 

    ceph config set osd bluestore_compression_min_blob_size SIZE
ceph config set osd bluestore_compression_max_blob_size SIZE
    Copy to Clipboard Toggle word wrap

    The following are the default blob sizes:

    SSD
    65536 bytes (minimum and maximum)
    HDD
    8192 bytes (minimum) and 65536 bytes (maximum)

5.1.3. Removing global compression settings
링크 복사

Remove global BlueStore compression settings to revert all OSDs to their default values. Removing these options affects only future writes. Existing data is not modified.

Procedure

  1. Remove the global compression algorithm. This command overrides any set algorithms so that OSDs stop forcing a specific compression algorithm. 

    ceph config rm osd bluestore_compression_algorithm
    Copy to Clipboard Toggle word wrap

  2. Remove the global compression mode. This command clears the compression mode (none, passive, aggressive, or force) and returns OSDs to the default mode. 

    ceph config rm osd bluestore_compression_mode
    Copy to Clipboard Toggle word wrap

  3. Remove the global required ratio. This command removes the threshold used to decide whether compressed data should be kept. 

    ceph config rm osd bluestore_compression_required_ratio
    Copy to Clipboard Toggle word wrap

  4. Remove the minimum blog size. This command removes the configured minimum blob size so that OSDs use the default value based on the device type. 

    ceph config rm osd bluestore_compression_min_blob_size
    Copy to Clipboard Toggle word wrap

  5. Remove the maximum blog size. This command removes the configured maximum blob size so that OSDs use the default value based on the device type. 

    ceph config rm osd bluestore_compression_max_blob_size
    Copy to Clipboard Toggle word wrap
Red Hat logoGithubredditYoutubeTwitter

자세한 정보

평가판, 구매 및 판매

커뮤니티

Red Hat 문서 정보

Red Hat을 사용하는 고객은 신뢰할 수 있는 콘텐츠가 포함된 제품과 서비스를 통해 혁신하고 목표를 달성할 수 있습니다. 최신 업데이트를 확인하세요.

보다 포괄적 수용을 위한 오픈 소스 용어 교체

Red Hat은 코드, 문서, 웹 속성에서 문제가 있는 언어를 교체하기 위해 최선을 다하고 있습니다. 자세한 내용은 다음을 참조하세요.Red Hat 블로그.

Red Hat 소개

Red Hat은 기업이 핵심 데이터 센터에서 네트워크 에지에 이르기까지 플랫폼과 환경 전반에서 더 쉽게 작업할 수 있도록 강화된 솔루션을 제공합니다.

Theme

© 2026 Red Hat맨 위로 이동