When using compression and deduplication:
- I/O performance to data with compression and deduplication is degraded.
- Approximately 10% of the effective pool capacity is used for metadata and garbage data by using capacity saving.
- In deduplication and compression, processing is performed per 8 KB. Therefore, if the block size of the file system is an integral multiple of 8 KB, capacity saving is likely to be effective.
- If the write workload rate is higher than garbage collection throughput, Cache Write Pending increases, causing performance degradation. Contact customer support to determine the garbage collection throughput for your configuration.
- Data written after the setting is changed will not be deduplicated and will only be compressed.
- Data already deduplicated before the setting is changed will not be rehydrated after the change.
The following table shows examples of the garbage collection performance for individual V-VOLs with capacity saving enabled. The values are the same for the compression setting and the deduplication and compression setting.
| I/O pattern | VSP One B85 |
|---|---|
| Random Write (kIOPS) (8 KB) | 77 |
| Sequential Write (GB/s) (256 KB) | 1.4 |
Capacity saving restrictions
While the capacity saving function is expected to lower the cost of purchasing drives, depending on the nature and access patterns of the stored data, the capacity saving effect might not be as expected as shown in the following table.
If update writes continue to occur beyond the capacity of garbage collection, the cache write pending rate increases, potentially impacting the entire system. Additionally, I/O performance on compressed and deduplicated data degrades.
| Data type, environment, or business | Capacity saving function | Description |
|---|---|---|
| Microsoft Office data | Deduplication and compression | There are many copies of the same file, therefore deduplication is effective. |
| Virtual Desktop Infrastructure (VDI) | Deduplication and compression | The presence of a cloned OS area makes deduplication very effective. |
| Database | Compression | Each block contains unique information, therefore deduplication is not effective. |
| Image, video, or other compressed or encrypted data | Disabled | The application compresses the data. |
| Data backup or data archive | Deduplication and compression | There is a deduplication effect between backup data. |
If you want to perform fast operations of logical data backup, archiving, or utilization of replicated data, take snapshot data by using Thin Image Advanced with DRS-VOLs. Thin Image Advanced enables the fast operations of taking snapshot data by using metadata with storage location information. This table shows examples of combinations of functions in software products applied to take snapshot data.
| Combination of functions in software products | Capacity saving function | Description |
|---|---|---|
| DRS-VOLs and Thin Image Advanced |
Deduplication and compression Deduplication and compression are effective for a volume and snapshot data of Thin Image Advanced. |
|