meta data for this page
  •  

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Next revision		Previous revision
vm:proxmox:ceph:compression [2021/01/28 18:31] – created niziakvm:proxmox:ceph:compression [2025/10/29 10:47] (current) – niziak
Line 3: Line 3:
 [[https://docs.ceph.com/en/latest/rados/configuration/bluestore-config-ref/#inline-compression]] [[https://docs.ceph.com/en/latest/rados/configuration/bluestore-config-ref/#inline-compression]]
  
 +[[https://www.redhat.com/en/blog/red-hat-ceph-storage-33-bluestore-compression-performance|BlueStore compression performance]]
  
 +===== get compression ratio =====
 +
 +<code bash>
 +ceph df detail
 +
 +--- POOLS ---
 +POOL             ID  PGS   STORED   (DATA)  (OMAP)  OBJECTS     USED   (DATA)   (OMAP)  %USED  MAX AVAIL  QUOTA OBJECTS  QUOTA BYTES  DIRTY  USED COMPR  UNDER COMPR
 +rbd               1  512  2.3 TiB  2.3 TiB  15 KiB  638.96k  6.4 TiB  6.4 TiB   44 KiB  59.17    1.5 TiB            N/A          N/A    N/A     562 GiB      1.1 TiB
 +</code>
 +
 +  * **USED COMPR**: The amount of space allocated for compressed data. This includes compressed data in addition to all of the space required for replication, allocation granularity, and erasure- coding overhead.
 +  * **UNDER COMPR**: The amount of data that has passed through compression (summed over all replicas) and that is worth storing in a compressed form.
 +
 +===== enable compression  =====
 +
 +<code bash>
 +ceph osd pool set {pool-name} compression_algorithm {algorithm}
 +</code>
 +
 +  * {pool-name}: The name of the pool you want to modify.
 +  * {algorithm}: The compression algorithm to use.
 +    * zstd: A modern and fast compression algorithm.
 +    * snappy: A good default with high performance and a decent compression ratio.
 +    * lz4: Very fast but with a lower compression ratio.
 +    * none: Disables compression for the poo
 +    
 +* Snappy — A fast and efficient compression algorithm developed by Google. It prioritizes speed over compression ratio. Snappy is used by default in Spark.
 +* LZ4 — A very fast compression algorithm that focuses on decompression speed. LZ4 provides a good balance between speed and compression ratio.
 +* ZStandard (ZStd) — A modern compression algorithm that provides a good compression ratio while still being pretty fast. ZStd offers compression ratios comparable to ZLib with faster compression/decompression speeds.
 +
 +  * tbd
 +    * none: Never compress data.
 +    * passive: Do not compress data unless the write operation has a compressible hint set.
 +    * aggressive: Do compress data unless the write operation has an incompressible hint set.
 +    * force: Try to compress data no matter what.
 +    
 +<code bash>
 +ceph osd pool set rbd compression_algorithm zstd
 +ceph osd pool set rbd compression_mode aggressive
 +</code>
 +
 +<code bash>
 +ceph osd pool get rbd compression_mode
 +ceph osd pool get rbd compression_algorithm
 +</code>
 +
 +===== detailed statistics =====
 +
 +Check compression ratios:
 +<code bash>
 +ceph daemon osd.2 perf dump bluestore | egrep -i "compressed_|allocated|stored|compress_.*_count"
 +ceph tell 'osd.*' perf dump bluestore | egrep -i "compressed_|allocated|stored|compress_.*_count"
 +</code>
 +
 +or
 <code bash> <code bash>
-ceph osd pool set rbd compresison_algorithm zstd +ceph daemon osd.0 perf dump bluestore | egrep -i "compressed_|allocated|stored|compress_.*_count|write_big|write_small" 
-ceph osd pool set rbd compression_mode <mode> +ceph tell 'osd.*' perf dump bluestore | egrep -i "compressed_|allocated|stored|compress_.*_count|write_big|write_small"
-ceph osd pool set <pool-name> compression_required_ratio <ratio>+
 </code> </code>