diff --git a/content/influxdb/v2/install/upgrade/v1-to-v2/manual-upgrade.md b/content/influxdb/v2/install/upgrade/v1-to-v2/manual-upgrade.md
index 6d2ee101af..fe0051d578 100644
--- a/content/influxdb/v2/install/upgrade/v1-to-v2/manual-upgrade.md
+++ b/content/influxdb/v2/install/upgrade/v1-to-v2/manual-upgrade.md
@@ -61,7 +61,7 @@ If you're using custom configuration settings in your InfluxDB 1.x instance, do
 | cache-snapshot-write-cold-duration | [storage-cache-snapshot-write-cold-duration](/influxdb/v2/reference/config-options/#storage-cache-snapshot-write-cold-duration) |
 | compact-full-write-cold-duration   | [storage-compact-full-write-cold-duration](/influxdb/v2/reference/config-options/#storage-compact-full-write-cold-duration)     |
 | max-concurrent-compactions         | [storage-max-concurrent-compactions](/influxdb/v2/reference/config-options/#storage-max-concurrent-compactions)                 |
-| compact-throughput                 |                                                                                                                                   |
+| compact-throughput                 | (not exposed as a separate option; see `storage-compact-throughput` notes)                                                        |
 | compact-throughput-burst           | [storage-compact-throughput-burst](/influxdb/v2/reference/config-options/#storage-compact-throughput-burst)                     |
 | tsm-use-madv-willneed              | [storage-tsm-use-madv-willneed](/influxdb/v2/reference/config-options/#storage-tsm-use-madv-willneed)                           |
 | max-series-per-database            |                                                                                                                                   |
diff --git a/content/influxdb/v2/reference/config-options.md b/content/influxdb/v2/reference/config-options.md
index 35c5e13374..9112652451 100644
--- a/content/influxdb/v2/reference/config-options.md
+++ b/content/influxdb/v2/reference/config-options.md
@@ -170,6 +170,7 @@ To configure InfluxDB, use the following configuration options when starting the
 - [storage-cache-snapshot-memory-size](#storage-cache-snapshot-memory-size)
 - [storage-cache-snapshot-write-cold-duration](#storage-cache-snapshot-write-cold-duration)
 - [storage-compact-full-write-cold-duration](#storage-compact-full-write-cold-duration)
+- [storage-compact-throughput](#storage-compact-throughput)
 - [storage-compact-throughput-burst](#storage-compact-throughput-burst)
 - [storage-max-concurrent-compactions](#storage-max-concurrent-compactions)
 - [storage-max-index-log-file-size](#storage-max-index-log-file-size)
@@ -2190,8 +2191,78 @@ storage-compact-full-write-cold-duration = "4h0m0s"
 
 ---
 
+### storage-compact-throughput
+Sustained rate limit (in bytes per second) that TSM compactions can write to disk.
+
+InfluxDB also supports a separate burst limit (`storage-compact-throughput-burst`).
+In InfluxDB OSS v2, the sustained and burst compaction throughput values are currently the same.
+To tune compaction write throughput, set `storage-compact-throughput-burst` and verify the effective values in the `Compaction settings` log entry at startup.
+
+Example log entry:
+
+```text
+Compaction settings {"max_concurrent_compactions": 7, "throughput_bytes_per_second": 50331648, "throughput_bytes_per_second_burst": 50331648}
+```
+
+**Default:** `50331648`
+
+| influxd flag | Environment variable | Configuration key |
+| :----------- | :------------------- | :---------------- |
+| _Not available_ | `INFLUXD_STORAGE_COMPACT_THROUGHPUT` | `storage-compact-throughput` |
+
+###### Environment variable
+```sh
+export INFLUXD_STORAGE_COMPACT_THROUGHPUT=50331648
+```
+
+###### Configuration file
+{{< code-tabs-wrapper >}}
+{{% code-tabs %}}
+[YAML](#)
+[TOML](#)
+[JSON](#)
+{{% /code-tabs %}}
+{{% code-tab-content %}}
+```yml
+storage-compact-throughput: 50331648
+```
+{{% /code-tab-content %}}
+{{% code-tab-content %}}
+```toml
+storage-compact-throughput = 50331648
+```
+{{% /code-tab-content %}}
+{{% code-tab-content %}}
+```json
+{
+  "storage-compact-throughput": 50331648
+}
+```
+{{% /code-tab-content %}}
+{{< /code-tabs-wrapper >}}
+
+---
+
 ### storage-compact-throughput-burst
-Rate limit (in bytes per second) that TSM compactions can write to disk.
+Maximum rate limit (in bytes per second) that TSM compactions can write to disk.
+
+In InfluxDB OSS v2, this setting also effectively controls the sustained compaction throughput.
+
+#### Scale compaction throughput
+
+If you scale to larger machines and faster storage, increase this value to let compactions keep up with ingest.
+If you set this too high, compactions can compete with writes and queries for disk I/O and hurt performance.
+
+To identify a compaction bottleneck, monitor the following:
+
+- `storage_compactions_queued` and `storage_compactions_active` from the `/metrics` endpoint
+- Disk I/O utilization and throughput from your OS or infrastructure metrics
+
+Example values:
+
+- `50331648` (default, about 48 MiB/s)
+- `100663296` (about 96 MiB/s)
+- `201326592` (about 192 MiB/s)
 
 **Default:** `50331648`
 
diff --git a/content/influxdb/v2/reference/internals/shards.md b/content/influxdb/v2/reference/internals/shards.md
index 3b5c93987b..25a0e1319b 100644
--- a/content/influxdb/v2/reference/internals/shards.md
+++ b/content/influxdb/v2/reference/internals/shards.md
@@ -134,7 +134,9 @@ The following configuration settings are especially beneficial for systems with
 - [`storage-max-concurrent-compactions`](/influxdb/v2/reference/config-options/#storage-max-concurrent-compactions)
 - [`storage-max-index-log-file-size`](/influxdb/v2/reference/config-options/#storage-max-index-log-file-size)
 - [`storage-series-file-max-concurrent-snapshot-compactions`](/influxdb/v2/reference/config-options/#storage-series-file-max-concurrent-snapshot-compactions)
-- [`storage-series-file-max-concurrent-snapshot-compactions`](/influxdb/v2/reference/config-options/#storage-series-file-max-concurrent-snapshot-compactions)
+
+In InfluxDB OSS v2, `storage-compact-throughput-burst` effectively controls both the sustained and burst compaction throughput.
+Use it to scale compaction write throughput as you scale CPU and disk.
 
 In systems with stable loads, if compactions interfere with other operations, typically, the system is undersized for its load, and configuration changes won't help much.