feat: more than one flow workers (#5315)

* feat: more workers

* feat: use round robin

* refactor: per review

* refactor: per bot review

* chore: per review

* docs: example

* docs: update config.md

* docs: update

* chore: per review

* refactor: set workers to cpu/2.max(1)

* fix: flow config in standalone mode

* test: fix config test

* docs: update docs&opt name

* chore: update config.md

* refactor: per review, sanitize at top

* chore: per review

* chore: config.md
This commit is contained in:
discord9
2025-01-14 20:28:18 +08:00
committed by GitHub
parent 95b20592ac
commit 66f63ae981
12 changed files with 120 additions and 30 deletions

View File

@@ -91,6 +91,8 @@
| `procedure` | -- | -- | Procedure storage options. |
| `procedure.max_retry_times` | Integer | `3` | Procedure max retry time. |
| `procedure.retry_delay` | String | `500ms` | Initial retry delay of procedures, increases exponentially |
| `flow` | -- | -- | flow engine options. |
| `flow.num_workers` | Integer | `0` | The number of flow worker in flownode.<br/>Not setting(or set to 0) this value will use the number of CPU cores divided by 2. |
| `storage` | -- | -- | The data storage options. |
| `storage.data_home` | String | `/tmp/greptimedb/` | The working home directory. |
| `storage.type` | String | `File` | The storage type used to store the data.<br/>- `File`: the data is stored in the local file system.<br/>- `S3`: the data is stored in the S3 object storage.<br/>- `Gcs`: the data is stored in the Google Cloud Storage.<br/>- `Azblob`: the data is stored in the Azure Blob Storage.<br/>- `Oss`: the data is stored in the Aliyun OSS. |
@@ -536,6 +538,8 @@
| --- | -----| ------- | ----------- |
| `mode` | String | `distributed` | The running mode of the flownode. It can be `standalone` or `distributed`. |
| `node_id` | Integer | Unset | The flownode identifier and should be unique in the cluster. |
| `flow` | -- | -- | flow engine options. |
| `flow.num_workers` | Integer | `0` | The number of flow worker in flownode.<br/>Not setting(or set to 0) this value will use the number of CPU cores divided by 2. |
| `grpc` | -- | -- | The gRPC server options. |
| `grpc.addr` | String | `127.0.0.1:6800` | The address to bind the gRPC server. |
| `grpc.hostname` | String | `127.0.0.1` | The hostname advertised to the metasrv,<br/>and used for connections from outside the host |

View File

@@ -5,6 +5,12 @@ mode = "distributed"
## @toml2docs:none-default
node_id = 14
## flow engine options.
[flow]
## The number of flow worker in flownode.
## Not setting(or set to 0) this value will use the number of CPU cores divided by 2.
#+num_workers=0
## The gRPC server options.
[grpc]
## The address to bind the gRPC server.

View File

@@ -284,6 +284,12 @@ max_retry_times = 3
## Initial retry delay of procedures, increases exponentially
retry_delay = "500ms"
## flow engine options.
[flow]
## The number of flow worker in flownode.
## Not setting(or set to 0) this value will use the number of CPU cores divided by 2.
#+num_workers=0
# Example of using S3 as the storage.
# [storage]
# type = "S3"