You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
- Use getLongProperty() instead of getProperty() to avoid native string
allocation in JNI layer during high-frequency Prometheus metrics scraping
- Add kv.rocksdb.statistics.enabled config option (default: false) to
allow disabling RocksDB statistics collection for reduced overhead
- Refactor getPropertyValue to getPropertyLongValue and eliminate code
duplication between overloaded methods
- Add configuration documentation for the new option
Copy file name to clipboardExpand all lines: website/docs/maintenance/configuration.md
+1Lines changed: 1 addition & 0 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -162,6 +162,7 @@ during the Fluss cluster working.
162
162
| kv.rocksdb.bloom-filter.bits-per-key | Double | 10.0 | Bits per key that bloom filter will use, this only take effect when bloom filter is used. The default value is 10.0. |
163
163
| kv.rocksdb.bloom-filter.block-based-mode | Boolean | false | If true, RocksDB will use block-based filter instead of full filter, this only take effect when bloom filter is used. The default value is `false`. |
164
164
| kv.rocksdb.shared-rate-limiter-bytes-per-sec | MemorySize | Long.MAX_VALUE | The bytes per second rate limit for RocksDB flush and compaction operations shared across all RocksDB instances on the TabletServer. The rate limiter is always enabled. The default value is Long.MAX_VALUE (effectively unlimited). Set to a lower value (e.g., 100MB) to limit the rate. This configuration can be updated dynamically without server restart. See [Updating Configs](operations/updating-configs.md) for more details. |
165
+
| kv.rocksdb.statistics.enabled | Boolean | true | Whether to enable RocksDB statistics collection for metrics. When enabled, RocksDB will collect various statistics like bytes read/written, compaction time, flush time, etc., which can be exposed through Fluss metrics. Enabling statistics has a small performance overhead (typically < 5%). If you experience performance issues or don't need RocksDB-level metrics, you can disable this option to reduce overhead. |
165
166
| kv.recover.log-record-batch.max-size | MemorySize | 16mb | The max fetch size for fetching log to apply to kv during recovering kv. |
0 commit comments