This is an automated email from the ASF dual-hosted git repository.
kassiez pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/doris-website.git
The following commit(s) were added to refs/heads/master by this push:
new 174da72fc67 Installation optimization (#1661)
174da72fc67 is described below
commit 174da72fc67d8dddb75edc5ff7f3789098181787
Author: wangtianyi2004 <376612...@qq.com>
AuthorDate: Thu Jan 2 14:16:49 2025 +0800
Installation optimization (#1661)
## Versions
- [ ] dev
- [ ] 3.0
- [ ] 2.1
- [ ] 2.0
## Languages
- [ ] Chinese
- [ ] English
## Docs Checklist
- [ ] Checked by AI
- [ ] Test Cases Built
---
.../storage-compute-coupled-deploy-manually.md | 2 --
docs/install/preparation/cluster-planning.md | 1 -
docs/install/preparation/env-checking.md | 35 ++++++++++++++++++++--
.../storage-compute-coupled-deploy-manually.md | 4 +--
.../storage-compute-decoupled-deploy-manually.md | 4 ++-
.../install/preparation/cluster-planning.md | 1 -
.../current/install/preparation/env-checking.md | 29 +++++++++++++++++-
.../current/install/preparation/os-checking.md | 2 +-
.../storage-compute-coupled-deploy-manually.md | 4 +--
.../install/preparation/cluster-planning.md | 1 -
.../install/preparation/env-checking.md | 29 +++++++++++++++++-
.../version-2.1/install/preparation/os-checking.md | 2 +-
.../storage-compute-coupled-deploy-manually.md | 4 +--
.../storage-compute-decoupled-deploy-manually.md | 4 ++-
.../install/preparation/cluster-planning.md | 1 -
.../install/preparation/env-checking.md | 29 +++++++++++++++++-
.../version-3.0/install/preparation/os-checking.md | 2 +-
.../install/preparation/cluster-planning.md | 1 -
.../install/preparation/env-checking.md | 35 ++++++++++++++++++++--
.../install/preparation/cluster-planning.md | 1 -
.../install/preparation/env-checking.md | 35 ++++++++++++++++++++--
21 files changed, 198 insertions(+), 28 deletions(-)
diff --git
a/docs/install/deploy-manually/storage-compute-coupled-deploy-manually.md
b/docs/install/deploy-manually/storage-compute-coupled-deploy-manually.md
index 7bc0748a6c9..66d8d0e1a0a 100644
--- a/docs/install/deploy-manually/storage-compute-coupled-deploy-manually.md
+++ b/docs/install/deploy-manually/storage-compute-coupled-deploy-manually.md
@@ -136,8 +136,6 @@ In a production cluster, it is recommended to deploy at
least 3 Follower nodes.
- The number of FE Follower nodes (including Master) should be odd. It is
recommended to deploy 3 nodes for high availability.
- When FE is deployed in high availability mode (1 Master, 2 Followers), we
recommend adding Observer FE nodes to extend the FE read service capacity.
-
- - Typically, one FE node can handle 10-20 BE nodes. It is recommended that
the total number of FE nodes be kept below 10.
:::
4. Start FE Follower Node
diff --git a/docs/install/preparation/cluster-planning.md
b/docs/install/preparation/cluster-planning.md
index fadd0764777..cb1ca6f7d80 100644
--- a/docs/install/preparation/cluster-planning.md
+++ b/docs/install/preparation/cluster-planning.md
@@ -39,7 +39,6 @@ Doris instances communicate over the network, and their
proper functioning requi
| FE | rpc_port | 9020 | BE -> FE, FE <-> FE
| Thrift Server port on FE, each FE should have the same
configuration |
| FE | query_port | 9030 | Client <-> FE
| MySQL Server port on FE |
| FE | edit_log_port | 9010 | FE <-> FE
| bdbje communication port on FE |
-| Broker | broker_ipc_port | 8000 | FE -> Broker, BE ->
Broker | Thrift Server port on Broker, used to receive requests from FE and
BE |
## Node Count Planning
diff --git a/docs/install/preparation/env-checking.md
b/docs/install/preparation/env-checking.md
index 6169263e471..c9dba1f7bbf 100644
--- a/docs/install/preparation/env-checking.md
+++ b/docs/install/preparation/env-checking.md
@@ -91,12 +91,43 @@ Doris involves distributing data partitions across
different instances for paral
Doris can be deployed on x86-64 or ARM64 architecture server platforms.
- Development and Testing Environments
-
+
+ | Module | CPU | Memory | Disk | Network
| Minimum Instance Count |
+ | --------- | ---------- | ------- | ---------------------------- |
-------------------- | ---------------------- |
+ | Frontend | 8 cores + | 8 GB+ | SSD or SATA, 10 GB+ |
Gigabit/Ten-Gigabit | 1 |
+ | Backend | 8 cores + | 16 GB+ | SSD or SATA, 50 GB+ |
Gigabit/Ten-Gigabit | 1 |
+
+ :::info
+
+ * In the test validation environment, you can deploy FE and BE on the same
server.
+ * It is generally recommended to deploy only one BE instance per machine
and only one FE instance.
+ * If 3 replicas of data are required, at least 3 machines should each
deploy one BE instance, instead of deploying 3 instances on a single machine.
+
+ :::
+
- Production Environments
+ | Module | CPU | Memory | Disk | Network
| Minimum Instance Count |
+ | --------- | ---------- | -------- | ---------------------------- |
--------- | ---------------------- |
+ | Frontend | 16 cores + | 64 GB+ | SSD or RAID card, 100 GB+ |
10-Gigabit | 1 |
+ | Backend | 16 cores + | 64 GB+ | SSD or SATA, 100 GB+ |
10-Gigabit | 3 |
+
+ :::info
+
+ * In production environments, if FE and BE are mixed, attention should be
paid to resource contention. It is recommended to store metadata and data on
separate disks.
+ * BE nodes can be configured with multiple hard drives, binding multiple
HDDs or SSDs on a single BE instance.
+ * The performance of the cluster depends on the resources of the BE nodes.
The more BE nodes, the better the Doris performance. Typically, Doris can fully
perform on clusters with 10 to 100 machines.
+
+ :::
+
## Disk Space Calculation
-In the Doris cluster, FE (Frontend) is mainly used for metadata storage,
including metadata edit logs and images. BE (Backend) disk space is primarily
used for storing data, and it needs to be calculated based on business
requirements. Starting from version 2.0, the use of the Broker component for
data import is no longer recommended.
+In the Doris cluster, FE (Frontend) is mainly used for metadata storage,
including metadata edit logs and images. BE (Backend) disk space is primarily
used for storing data, and it needs to be calculated based on business
requirements.
+| Component | Disk Space Description
|
+| --------- |
------------------------------------------------------------------------------------------------------------------
|
+| FE | It is recommended to reserve more than 100 GB of storage space,
using SSD disks. |
+| BE | Doris uses LZ4 compression by default. The compression ratio is
around 0.3 - 0.5. Disk space should be calculated as total data volume * 3 (for
3 replicas), and 40% of the space should be reserved for backend compaction and
temporary data storage. |
+
## Java Environment Check
diff --git
a/i18n/zh-CN/docusaurus-plugin-content-docs/current/install/deploy-manually/storage-compute-coupled-deploy-manually.md
b/i18n/zh-CN/docusaurus-plugin-content-docs/current/install/deploy-manually/storage-compute-coupled-deploy-manually.md
index d98d4819129..3976195fae9 100644
---
a/i18n/zh-CN/docusaurus-plugin-content-docs/current/install/deploy-manually/storage-compute-coupled-deploy-manually.md
+++
b/i18n/zh-CN/docusaurus-plugin-content-docs/current/install/deploy-manually/storage-compute-coupled-deploy-manually.md
@@ -32,6 +32,8 @@ under the License.
4. 验证集群正确性：部署完成后连接并验证集群正确性。
+在开始部署操作前，可以[下载](https://doris.apache.org/download)相应的 Doris 版本。
+
## 第 1 步：部署 FE Master 节点
1. 创建元数据路径
@@ -138,8 +140,6 @@ under the License.
- FE Follower（包括 Master）节点的数量建议为奇数，建议部署 3 个组成高可用模式。
- 当 FE 处于高可用部署时（1 个 Master，2 个 Follower），我们建议通过增加 Observer FE 来扩展 FE 的读服务能力
-
- - 通常一个 FE 节点可以应对 10-20 台 BE 节点。建议总的 FE 节点数量在 10 个以下
:::
4. 启动 FE Follower 节点
diff --git
a/i18n/zh-CN/docusaurus-plugin-content-docs/current/install/deploy-manually/storage-compute-decoupled-deploy-manually.md
b/i18n/zh-CN/docusaurus-plugin-content-docs/current/install/deploy-manually/storage-compute-decoupled-deploy-manually.md
index 5a4ca64802b..a23d60a0920 100644
---
a/i18n/zh-CN/docusaurus-plugin-content-docs/current/install/deploy-manually/storage-compute-decoupled-deploy-manually.md
+++
b/i18n/zh-CN/docusaurus-plugin-content-docs/current/install/deploy-manually/storage-compute-decoupled-deploy-manually.md
@@ -40,6 +40,8 @@ under the License.
9. 添加 Storage Vault：使用共享存储创建一个或多个 Storage Vault。
+在开始部署操作前，可以[下载](https://doris.apache.org/download)相应的 Doris 版本。
+
## 第 1 步：准备 FoundationDB
本节提供了脚本 `fdb_vars.sh` 和 `fdb_ctl.sh` 配置、部署和启动 FDB（FoundationDB）服务的分步指南。您可以下载
[doris
tools](http://apache-doris-releases.oss-accelerate.aliyuncs.com/apache-doris-3.0.2-tools.tar.gz)
并从 `fdb` 目录获取 `fdb_vars.sh` 和 `fdb_ctl.sh`。
@@ -174,7 +176,7 @@ Apache Doris 存算分离模式会将数据存储在 S3 服务或 HDFS 服务上
## 第 4 步：数据回收功能独立部署（可选）
-：::info 信息
+::info 信息
Meta Service 本身具备了元数据管理和回收功能，这两个功能可以独立部署，如果你想独立部署，可以参考这一节。
diff --git
a/i18n/zh-CN/docusaurus-plugin-content-docs/current/install/preparation/cluster-planning.md
b/i18n/zh-CN/docusaurus-plugin-content-docs/current/install/preparation/cluster-planning.md
index 34e964759d0..c779ca2cecc 100644
---
a/i18n/zh-CN/docusaurus-plugin-content-docs/current/install/preparation/cluster-planning.md
+++
b/i18n/zh-CN/docusaurus-plugin-content-docs/current/install/preparation/cluster-planning.md
@@ -39,7 +39,6 @@ Doris 的各个实例通过网络进行通信，其正常运行需要网络环
| FE | rpc_port | 9020 | BE -> FE，FE <-> FE |
FE 上的 Thrift Server 端口，每个 FE 的配置需保持一致 |
| FE | query_port | 9030 | Client <-> FE |
FE 上的 MySQL Server 端口 |
| FE | edit_log_port | 9010 | FE <-> FE |
FE 上的 bdbje 通信端口 |
-| Broker | broker_ipc_port | 8000 | FE -> Broker，BE -> Broker |
Broker 上的 Thrift Server 端口，用于接收请求 |
## 节点数量规划
diff --git
a/i18n/zh-CN/docusaurus-plugin-content-docs/current/install/preparation/env-checking.md
b/i18n/zh-CN/docusaurus-plugin-content-docs/current/install/preparation/env-checking.md
index 03b0a42908f..a36c1d45b78 100644
---
a/i18n/zh-CN/docusaurus-plugin-content-docs/current/install/preparation/env-checking.md
+++
b/i18n/zh-CN/docusaurus-plugin-content-docs/current/install/preparation/env-checking.md
@@ -91,11 +91,38 @@ Doris 支持运行和部署在 x86-64 架构的服务器平台或 ARM64 架构
- 开发及测试环境
+ | 模块 | CPU | 内存 | 磁盘 | 网络 | 实例数量（最低要求） |
+ | -------- | ----- | ------ | ----------------- | ------- | ---------- |
+ | Frontend | 8 核 + | 8 GB+ | SSD 或 SATA，10 GB+ | 千兆/万兆网卡 | 1 |
+ | Backend | 8 核 + | 16 GB+ | SSD 或 SATA，50 GB+ | 千兆/万兆网卡 | 1 |
+
+ :::info 说明
+ * 在验证测试环境中，可以将 FE 与 BE 部署在同一台服务器上
+ * 一台机器上一般只建议部署一个 BE 实例，同时只能部署一个 FE
+ * 如果需要 3 副本数据，那么至少需要 3 台机器各部署一个 BE 实例，而不是 1 台机器部署 3
+ :::
+
- 生产环境
+
+ | 模块 | CPU | 内存 | 磁盘 | 网络 | 实例数量（最低要求） |
+ | -------- | ------ | ------ | ------------------- | ---- | ---------- |
+ | Frontend | 16 核 + | 64 GB+ | SSD 或 RAID 卡，100GB+ | 万兆网卡 | 1 |
+ | Backend | 16 核 + | 64 GB+ | SSD 或 SATA，100G+ | 万兆网卡 | 3 |
+
+ :::info 说明
+ * 在生产环境中，如果 FE 与 BE 混布，需要注意资源争用问题，建议元数据存储与数据存储分盘存放
+ * BE 节点可以配置多块硬盘存储，在一个 BE 实例上绑定多块 HDD 或 SSD 盘
+ * 集群的性能与 BE 节点的资源有关，BE 节点越多，Doris 性能越好。通常情况下在 10 - 100 台机器上可以充分发挥 Doris 的性能
+ :::
## 硬盘空间计算
-在 Doris 集群中，FE 主要用于元数据存储，包括元数据 edit log 和 image。BE 的磁盘空间主要用于存放数据，需要根据业务需求计算。在
2.0 版本后，不再建议使用 Broker 组件进行数据导入。
+在 Doris 集群中，FE 主要用于元数据存储，包括元数据 edit log 和 image。BE 的磁盘空间主要用于存放数据，需要根据业务需求计算。
+
+| 组件 | 磁盘空间说明
|
+| ------ |
-----------------------------------------------------------------------------------------------------
|
+| FE | 建议预留 100GB 以上的存储空间，使用 SSD 硬盘。
|
+| BE | Doris 默认 LZ4 压缩方式进行存储，压缩比在 0.3 - 0.5 左右磁盘空间需要按照总数据量 \* 3（3
副本）计算需要预留出 40% 空间用作后台 compaction 以及临时数据的存储 |
## Java 环境检查
diff --git
a/i18n/zh-CN/docusaurus-plugin-content-docs/current/install/preparation/os-checking.md
b/i18n/zh-CN/docusaurus-plugin-content-docs/current/install/preparation/os-checking.md
index 4d97d1810b1..161dd6285b7 100644
---
a/i18n/zh-CN/docusaurus-plugin-content-docs/current/install/preparation/os-checking.md
+++
b/i18n/zh-CN/docusaurus-plugin-content-docs/current/install/preparation/os-checking.md
@@ -88,7 +88,7 @@ chmod +x /etc/rc.d/rc.local
## 确保系统有足够大的虚拟内存区域
-为了保证 Doris 有足够的内存映射区域来处理大量数据，需要修改虚拟内存区域的大小。如果没有足够的内存映射区域，Doris 在启动或运行时可能会遇到
`Too many open files` 或类似的错误。
+为了保证 Doris 有足够的内存映射区域来处理大量数据，需要修改 VMA（虚拟内存区域）。如果没有足够的内存映射区域，Doris 在启动或运行时可能会遇到
`Too many open files` 或类似的错误。
通过以下命令可以永久修改虚拟内存区域至少为 2000000，并立即生效：
diff --git
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-2.1/install/deploy-manually/storage-compute-coupled-deploy-manually.md
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-2.1/install/deploy-manually/storage-compute-coupled-deploy-manually.md
index d98d4819129..3976195fae9 100644
---
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-2.1/install/deploy-manually/storage-compute-coupled-deploy-manually.md
+++
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-2.1/install/deploy-manually/storage-compute-coupled-deploy-manually.md
@@ -32,6 +32,8 @@ under the License.
4. 验证集群正确性：部署完成后连接并验证集群正确性。
+在开始部署操作前，可以[下载](https://doris.apache.org/download)相应的 Doris 版本。
+
## 第 1 步：部署 FE Master 节点
1. 创建元数据路径
@@ -138,8 +140,6 @@ under the License.
- FE Follower（包括 Master）节点的数量建议为奇数，建议部署 3 个组成高可用模式。
- 当 FE 处于高可用部署时（1 个 Master，2 个 Follower），我们建议通过增加 Observer FE 来扩展 FE 的读服务能力
-
- - 通常一个 FE 节点可以应对 10-20 台 BE 节点。建议总的 FE 节点数量在 10 个以下
:::
4. 启动 FE Follower 节点
diff --git
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-2.1/install/preparation/cluster-planning.md
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-2.1/install/preparation/cluster-planning.md
index 34e964759d0..c779ca2cecc 100644
---
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-2.1/install/preparation/cluster-planning.md
+++
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-2.1/install/preparation/cluster-planning.md
@@ -39,7 +39,6 @@ Doris 的各个实例通过网络进行通信，其正常运行需要网络环
| FE | rpc_port | 9020 | BE -> FE，FE <-> FE |
FE 上的 Thrift Server 端口，每个 FE 的配置需保持一致 |
| FE | query_port | 9030 | Client <-> FE |
FE 上的 MySQL Server 端口 |
| FE | edit_log_port | 9010 | FE <-> FE |
FE 上的 bdbje 通信端口 |
-| Broker | broker_ipc_port | 8000 | FE -> Broker，BE -> Broker |
Broker 上的 Thrift Server 端口，用于接收请求 |
## 节点数量规划
diff --git
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-2.1/install/preparation/env-checking.md
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-2.1/install/preparation/env-checking.md
index 03b0a42908f..a36c1d45b78 100644
---
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-2.1/install/preparation/env-checking.md
+++
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-2.1/install/preparation/env-checking.md
@@ -91,11 +91,38 @@ Doris 支持运行和部署在 x86-64 架构的服务器平台或 ARM64 架构
- 开发及测试环境
+ | 模块 | CPU | 内存 | 磁盘 | 网络 | 实例数量（最低要求） |
+ | -------- | ----- | ------ | ----------------- | ------- | ---------- |
+ | Frontend | 8 核 + | 8 GB+ | SSD 或 SATA，10 GB+ | 千兆/万兆网卡 | 1 |
+ | Backend | 8 核 + | 16 GB+ | SSD 或 SATA，50 GB+ | 千兆/万兆网卡 | 1 |
+
+ :::info 说明
+ * 在验证测试环境中，可以将 FE 与 BE 部署在同一台服务器上
+ * 一台机器上一般只建议部署一个 BE 实例，同时只能部署一个 FE
+ * 如果需要 3 副本数据，那么至少需要 3 台机器各部署一个 BE 实例，而不是 1 台机器部署 3
+ :::
+
- 生产环境
+
+ | 模块 | CPU | 内存 | 磁盘 | 网络 | 实例数量（最低要求） |
+ | -------- | ------ | ------ | ------------------- | ---- | ---------- |
+ | Frontend | 16 核 + | 64 GB+ | SSD 或 RAID 卡，100GB+ | 万兆网卡 | 1 |
+ | Backend | 16 核 + | 64 GB+ | SSD 或 SATA，100G+ | 万兆网卡 | 3 |
+
+ :::info 说明
+ * 在生产环境中，如果 FE 与 BE 混布，需要注意资源争用问题，建议元数据存储与数据存储分盘存放
+ * BE 节点可以配置多块硬盘存储，在一个 BE 实例上绑定多块 HDD 或 SSD 盘
+ * 集群的性能与 BE 节点的资源有关，BE 节点越多，Doris 性能越好。通常情况下在 10 - 100 台机器上可以充分发挥 Doris 的性能
+ :::
## 硬盘空间计算
-在 Doris 集群中，FE 主要用于元数据存储，包括元数据 edit log 和 image。BE 的磁盘空间主要用于存放数据，需要根据业务需求计算。在
2.0 版本后，不再建议使用 Broker 组件进行数据导入。
+在 Doris 集群中，FE 主要用于元数据存储，包括元数据 edit log 和 image。BE 的磁盘空间主要用于存放数据，需要根据业务需求计算。
+
+| 组件 | 磁盘空间说明
|
+| ------ |
-----------------------------------------------------------------------------------------------------
|
+| FE | 建议预留 100GB 以上的存储空间，使用 SSD 硬盘。
|
+| BE | Doris 默认 LZ4 压缩方式进行存储，压缩比在 0.3 - 0.5 左右磁盘空间需要按照总数据量 \* 3（3
副本）计算需要预留出 40% 空间用作后台 compaction 以及临时数据的存储 |
## Java 环境检查
diff --git
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-2.1/install/preparation/os-checking.md
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-2.1/install/preparation/os-checking.md
index 4d97d1810b1..161dd6285b7 100644
---
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-2.1/install/preparation/os-checking.md
+++
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-2.1/install/preparation/os-checking.md
@@ -88,7 +88,7 @@ chmod +x /etc/rc.d/rc.local
## 确保系统有足够大的虚拟内存区域
-为了保证 Doris 有足够的内存映射区域来处理大量数据，需要修改虚拟内存区域的大小。如果没有足够的内存映射区域，Doris 在启动或运行时可能会遇到
`Too many open files` 或类似的错误。
+为了保证 Doris 有足够的内存映射区域来处理大量数据，需要修改 VMA（虚拟内存区域）。如果没有足够的内存映射区域，Doris 在启动或运行时可能会遇到
`Too many open files` 或类似的错误。
通过以下命令可以永久修改虚拟内存区域至少为 2000000，并立即生效：
diff --git
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-3.0/install/deploy-manually/storage-compute-coupled-deploy-manually.md
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-3.0/install/deploy-manually/storage-compute-coupled-deploy-manually.md
index d98d4819129..3976195fae9 100644
---
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-3.0/install/deploy-manually/storage-compute-coupled-deploy-manually.md
+++
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-3.0/install/deploy-manually/storage-compute-coupled-deploy-manually.md
@@ -32,6 +32,8 @@ under the License.
4. 验证集群正确性：部署完成后连接并验证集群正确性。
+在开始部署操作前，可以[下载](https://doris.apache.org/download)相应的 Doris 版本。
+
## 第 1 步：部署 FE Master 节点
1. 创建元数据路径
@@ -138,8 +140,6 @@ under the License.
- FE Follower（包括 Master）节点的数量建议为奇数，建议部署 3 个组成高可用模式。
- 当 FE 处于高可用部署时（1 个 Master，2 个 Follower），我们建议通过增加 Observer FE 来扩展 FE 的读服务能力
-
- - 通常一个 FE 节点可以应对 10-20 台 BE 节点。建议总的 FE 节点数量在 10 个以下
:::
4. 启动 FE Follower 节点
diff --git
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-3.0/install/deploy-manually/storage-compute-decoupled-deploy-manually.md
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-3.0/install/deploy-manually/storage-compute-decoupled-deploy-manually.md
index 5a4ca64802b..a23d60a0920 100644
---
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-3.0/install/deploy-manually/storage-compute-decoupled-deploy-manually.md
+++
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-3.0/install/deploy-manually/storage-compute-decoupled-deploy-manually.md
@@ -40,6 +40,8 @@ under the License.
9. 添加 Storage Vault：使用共享存储创建一个或多个 Storage Vault。
+在开始部署操作前，可以[下载](https://doris.apache.org/download)相应的 Doris 版本。
+
## 第 1 步：准备 FoundationDB
本节提供了脚本 `fdb_vars.sh` 和 `fdb_ctl.sh` 配置、部署和启动 FDB（FoundationDB）服务的分步指南。您可以下载
[doris
tools](http://apache-doris-releases.oss-accelerate.aliyuncs.com/apache-doris-3.0.2-tools.tar.gz)
并从 `fdb` 目录获取 `fdb_vars.sh` 和 `fdb_ctl.sh`。
@@ -174,7 +176,7 @@ Apache Doris 存算分离模式会将数据存储在 S3 服务或 HDFS 服务上
## 第 4 步：数据回收功能独立部署（可选）
-：::info 信息
+::info 信息
Meta Service 本身具备了元数据管理和回收功能，这两个功能可以独立部署，如果你想独立部署，可以参考这一节。
diff --git
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-3.0/install/preparation/cluster-planning.md
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-3.0/install/preparation/cluster-planning.md
index 34e964759d0..c779ca2cecc 100644
---
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-3.0/install/preparation/cluster-planning.md
+++
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-3.0/install/preparation/cluster-planning.md
@@ -39,7 +39,6 @@ Doris 的各个实例通过网络进行通信，其正常运行需要网络环
| FE | rpc_port | 9020 | BE -> FE，FE <-> FE |
FE 上的 Thrift Server 端口，每个 FE 的配置需保持一致 |
| FE | query_port | 9030 | Client <-> FE |
FE 上的 MySQL Server 端口 |
| FE | edit_log_port | 9010 | FE <-> FE |
FE 上的 bdbje 通信端口 |
-| Broker | broker_ipc_port | 8000 | FE -> Broker，BE -> Broker |
Broker 上的 Thrift Server 端口，用于接收请求 |
## 节点数量规划
diff --git
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-3.0/install/preparation/env-checking.md
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-3.0/install/preparation/env-checking.md
index 03b0a42908f..a36c1d45b78 100644
---
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-3.0/install/preparation/env-checking.md
+++
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-3.0/install/preparation/env-checking.md
@@ -91,11 +91,38 @@ Doris 支持运行和部署在 x86-64 架构的服务器平台或 ARM64 架构
- 开发及测试环境
+ | 模块 | CPU | 内存 | 磁盘 | 网络 | 实例数量（最低要求） |
+ | -------- | ----- | ------ | ----------------- | ------- | ---------- |
+ | Frontend | 8 核 + | 8 GB+ | SSD 或 SATA，10 GB+ | 千兆/万兆网卡 | 1 |
+ | Backend | 8 核 + | 16 GB+ | SSD 或 SATA，50 GB+ | 千兆/万兆网卡 | 1 |
+
+ :::info 说明
+ * 在验证测试环境中，可以将 FE 与 BE 部署在同一台服务器上
+ * 一台机器上一般只建议部署一个 BE 实例，同时只能部署一个 FE
+ * 如果需要 3 副本数据，那么至少需要 3 台机器各部署一个 BE 实例，而不是 1 台机器部署 3
+ :::
+
- 生产环境
+
+ | 模块 | CPU | 内存 | 磁盘 | 网络 | 实例数量（最低要求） |
+ | -------- | ------ | ------ | ------------------- | ---- | ---------- |
+ | Frontend | 16 核 + | 64 GB+ | SSD 或 RAID 卡，100GB+ | 万兆网卡 | 1 |
+ | Backend | 16 核 + | 64 GB+ | SSD 或 SATA，100G+ | 万兆网卡 | 3 |
+
+ :::info 说明
+ * 在生产环境中，如果 FE 与 BE 混布，需要注意资源争用问题，建议元数据存储与数据存储分盘存放
+ * BE 节点可以配置多块硬盘存储，在一个 BE 实例上绑定多块 HDD 或 SSD 盘
+ * 集群的性能与 BE 节点的资源有关，BE 节点越多，Doris 性能越好。通常情况下在 10 - 100 台机器上可以充分发挥 Doris 的性能
+ :::
## 硬盘空间计算
-在 Doris 集群中，FE 主要用于元数据存储，包括元数据 edit log 和 image。BE 的磁盘空间主要用于存放数据，需要根据业务需求计算。在
2.0 版本后，不再建议使用 Broker 组件进行数据导入。
+在 Doris 集群中，FE 主要用于元数据存储，包括元数据 edit log 和 image。BE 的磁盘空间主要用于存放数据，需要根据业务需求计算。
+
+| 组件 | 磁盘空间说明
|
+| ------ |
-----------------------------------------------------------------------------------------------------
|
+| FE | 建议预留 100GB 以上的存储空间，使用 SSD 硬盘。
|
+| BE | Doris 默认 LZ4 压缩方式进行存储，压缩比在 0.3 - 0.5 左右磁盘空间需要按照总数据量 \* 3（3
副本）计算需要预留出 40% 空间用作后台 compaction 以及临时数据的存储 |
## Java 环境检查
diff --git
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-3.0/install/preparation/os-checking.md
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-3.0/install/preparation/os-checking.md
index 4d97d1810b1..161dd6285b7 100644
---
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-3.0/install/preparation/os-checking.md
+++
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-3.0/install/preparation/os-checking.md
@@ -88,7 +88,7 @@ chmod +x /etc/rc.d/rc.local
## 确保系统有足够大的虚拟内存区域
-为了保证 Doris 有足够的内存映射区域来处理大量数据，需要修改虚拟内存区域的大小。如果没有足够的内存映射区域，Doris 在启动或运行时可能会遇到
`Too many open files` 或类似的错误。
+为了保证 Doris 有足够的内存映射区域来处理大量数据，需要修改 VMA（虚拟内存区域）。如果没有足够的内存映射区域，Doris 在启动或运行时可能会遇到
`Too many open files` 或类似的错误。
通过以下命令可以永久修改虚拟内存区域至少为 2000000，并立即生效：
diff --git a/versioned_docs/version-2.1/install/preparation/cluster-planning.md
b/versioned_docs/version-2.1/install/preparation/cluster-planning.md
index fadd0764777..cb1ca6f7d80 100644
--- a/versioned_docs/version-2.1/install/preparation/cluster-planning.md
+++ b/versioned_docs/version-2.1/install/preparation/cluster-planning.md
@@ -39,7 +39,6 @@ Doris instances communicate over the network, and their
proper functioning requi
| FE | rpc_port | 9020 | BE -> FE, FE <-> FE
| Thrift Server port on FE, each FE should have the same
configuration |
| FE | query_port | 9030 | Client <-> FE
| MySQL Server port on FE |
| FE | edit_log_port | 9010 | FE <-> FE
| bdbje communication port on FE |
-| Broker | broker_ipc_port | 8000 | FE -> Broker, BE ->
Broker | Thrift Server port on Broker, used to receive requests from FE and
BE |
## Node Count Planning
diff --git a/versioned_docs/version-2.1/install/preparation/env-checking.md
b/versioned_docs/version-2.1/install/preparation/env-checking.md
index 6169263e471..c9dba1f7bbf 100644
--- a/versioned_docs/version-2.1/install/preparation/env-checking.md
+++ b/versioned_docs/version-2.1/install/preparation/env-checking.md
@@ -91,12 +91,43 @@ Doris involves distributing data partitions across
different instances for paral
Doris can be deployed on x86-64 or ARM64 architecture server platforms.
- Development and Testing Environments
-
+
+ | Module | CPU | Memory | Disk | Network
| Minimum Instance Count |
+ | --------- | ---------- | ------- | ---------------------------- |
-------------------- | ---------------------- |
+ | Frontend | 8 cores + | 8 GB+ | SSD or SATA, 10 GB+ |
Gigabit/Ten-Gigabit | 1 |
+ | Backend | 8 cores + | 16 GB+ | SSD or SATA, 50 GB+ |
Gigabit/Ten-Gigabit | 1 |
+
+ :::info
+
+ * In the test validation environment, you can deploy FE and BE on the same
server.
+ * It is generally recommended to deploy only one BE instance per machine
and only one FE instance.
+ * If 3 replicas of data are required, at least 3 machines should each
deploy one BE instance, instead of deploying 3 instances on a single machine.
+
+ :::
+
- Production Environments
+ | Module | CPU | Memory | Disk | Network
| Minimum Instance Count |
+ | --------- | ---------- | -------- | ---------------------------- |
--------- | ---------------------- |
+ | Frontend | 16 cores + | 64 GB+ | SSD or RAID card, 100 GB+ |
10-Gigabit | 1 |
+ | Backend | 16 cores + | 64 GB+ | SSD or SATA, 100 GB+ |
10-Gigabit | 3 |
+
+ :::info
+
+ * In production environments, if FE and BE are mixed, attention should be
paid to resource contention. It is recommended to store metadata and data on
separate disks.
+ * BE nodes can be configured with multiple hard drives, binding multiple
HDDs or SSDs on a single BE instance.
+ * The performance of the cluster depends on the resources of the BE nodes.
The more BE nodes, the better the Doris performance. Typically, Doris can fully
perform on clusters with 10 to 100 machines.
+
+ :::
+
## Disk Space Calculation
-In the Doris cluster, FE (Frontend) is mainly used for metadata storage,
including metadata edit logs and images. BE (Backend) disk space is primarily
used for storing data, and it needs to be calculated based on business
requirements. Starting from version 2.0, the use of the Broker component for
data import is no longer recommended.
+In the Doris cluster, FE (Frontend) is mainly used for metadata storage,
including metadata edit logs and images. BE (Backend) disk space is primarily
used for storing data, and it needs to be calculated based on business
requirements.
+| Component | Disk Space Description
|
+| --------- |
------------------------------------------------------------------------------------------------------------------
|
+| FE | It is recommended to reserve more than 100 GB of storage space,
using SSD disks. |
+| BE | Doris uses LZ4 compression by default. The compression ratio is
around 0.3 - 0.5. Disk space should be calculated as total data volume * 3 (for
3 replicas), and 40% of the space should be reserved for backend compaction and
temporary data storage. |
+
## Java Environment Check
diff --git a/versioned_docs/version-3.0/install/preparation/cluster-planning.md
b/versioned_docs/version-3.0/install/preparation/cluster-planning.md
index fadd0764777..cb1ca6f7d80 100644
--- a/versioned_docs/version-3.0/install/preparation/cluster-planning.md
+++ b/versioned_docs/version-3.0/install/preparation/cluster-planning.md
@@ -39,7 +39,6 @@ Doris instances communicate over the network, and their
proper functioning requi
| FE | rpc_port | 9020 | BE -> FE, FE <-> FE
| Thrift Server port on FE, each FE should have the same
configuration |
| FE | query_port | 9030 | Client <-> FE
| MySQL Server port on FE |
| FE | edit_log_port | 9010 | FE <-> FE
| bdbje communication port on FE |
-| Broker | broker_ipc_port | 8000 | FE -> Broker, BE ->
Broker | Thrift Server port on Broker, used to receive requests from FE and
BE |
## Node Count Planning
diff --git a/versioned_docs/version-3.0/install/preparation/env-checking.md
b/versioned_docs/version-3.0/install/preparation/env-checking.md
index 6169263e471..c9dba1f7bbf 100644
--- a/versioned_docs/version-3.0/install/preparation/env-checking.md
+++ b/versioned_docs/version-3.0/install/preparation/env-checking.md
@@ -91,12 +91,43 @@ Doris involves distributing data partitions across
different instances for paral
Doris can be deployed on x86-64 or ARM64 architecture server platforms.
- Development and Testing Environments
-
+
+ | Module | CPU | Memory | Disk | Network
| Minimum Instance Count |
+ | --------- | ---------- | ------- | ---------------------------- |
-------------------- | ---------------------- |
+ | Frontend | 8 cores + | 8 GB+ | SSD or SATA, 10 GB+ |
Gigabit/Ten-Gigabit | 1 |
+ | Backend | 8 cores + | 16 GB+ | SSD or SATA, 50 GB+ |
Gigabit/Ten-Gigabit | 1 |
+
+ :::info
+
+ * In the test validation environment, you can deploy FE and BE on the same
server.
+ * It is generally recommended to deploy only one BE instance per machine
and only one FE instance.
+ * If 3 replicas of data are required, at least 3 machines should each
deploy one BE instance, instead of deploying 3 instances on a single machine.
+
+ :::
+
- Production Environments
+ | Module | CPU | Memory | Disk | Network
| Minimum Instance Count |
+ | --------- | ---------- | -------- | ---------------------------- |
--------- | ---------------------- |
+ | Frontend | 16 cores + | 64 GB+ | SSD or RAID card, 100 GB+ |
10-Gigabit | 1 |
+ | Backend | 16 cores + | 64 GB+ | SSD or SATA, 100 GB+ |
10-Gigabit | 3 |
+
+ :::info
+
+ * In production environments, if FE and BE are mixed, attention should be
paid to resource contention. It is recommended to store metadata and data on
separate disks.
+ * BE nodes can be configured with multiple hard drives, binding multiple
HDDs or SSDs on a single BE instance.
+ * The performance of the cluster depends on the resources of the BE nodes.
The more BE nodes, the better the Doris performance. Typically, Doris can fully
perform on clusters with 10 to 100 machines.
+
+ :::
+
## Disk Space Calculation
-In the Doris cluster, FE (Frontend) is mainly used for metadata storage,
including metadata edit logs and images. BE (Backend) disk space is primarily
used for storing data, and it needs to be calculated based on business
requirements. Starting from version 2.0, the use of the Broker component for
data import is no longer recommended.
+In the Doris cluster, FE (Frontend) is mainly used for metadata storage,
including metadata edit logs and images. BE (Backend) disk space is primarily
used for storing data, and it needs to be calculated based on business
requirements.
+| Component | Disk Space Description
|
+| --------- |
------------------------------------------------------------------------------------------------------------------
|
+| FE | It is recommended to reserve more than 100 GB of storage space,
using SSD disks. |
+| BE | Doris uses LZ4 compression by default. The compression ratio is
around 0.3 - 0.5. Disk space should be calculated as total data volume * 3 (for
3 replicas), and 40% of the space should be reserved for backend compaction and
temporary data storage. |
+
## Java Environment Check
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@doris.apache.org
For additional commands, e-mail: commits-h...@doris.apache.org
