add attu docs, update readme

Signed-off-by: nameczz <zizhao.chen@zilliz.com>

README.md

@@ -8,6 +8,11 @@ Attu provides an intuitive and efficient GUI for Milvus, allowing you to interac
 
 <img src="./.github/images/screenshot.png" alt="attu" />
 
+## Attu Doc
+
+- [English](./doc/en.md)
+- [中文](./doc/zh-cn.md)
+
 ## Features and Roadmap
 
 Attu is under rapid development and we are adding new features weekly, here are the current plan, we will release a version once a feature is available.
@@ -40,10 +45,10 @@ Once you start the docker, open the browser, type `http://{ attu ip }:8000`, you
 
 #### Params
 
-| Parameter  | Example                 | required | description                                 |
-| :--------- | :---------------------- | :------: | ------------------------------------------- |
+| Parameter  | Example                 | required | description                       |
+| :--------- | :---------------------- | :------: | --------------------------------- |
 | HOST_URL   | http://192.168.0.1:8000 |   true   | Where attu container is installed |
-| MILVUS_URL | 192.168.0.1:19530       |  false   | Optional, Milvus server URL                 |
+| MILVUS_URL | 192.168.0.1:19530       |  false   | Optional, Milvus server URL       |
 
 Tip: **127.0.0.1 or localhost will not work when runs on docker**
 
@@ -84,8 +89,6 @@ New to milvus? Milvus is an open-source vector database built to power AI applic
 - [Milvus python sdk](https://milvus.io/docs/v2.0.0/explore_pymilvus.md)
 - [Milvus bootcamp](https://milvus.io/bootcamp)
 
-
-
 #### ❓ Questions? Problems?
 
 - If you've found a bug or want to request a feature, please create a [GitHub Issue](https://github.com/zilliztech/attu/issues/new/choose).

doc/en.md

@@ -0,0 +1,24 @@
+# Attu doc
+
+### Installation
+
+- [Run Attu in docker](./en/attu_install-docker.md)
+- [Run Attu in helm](./en/attu_install-helm.md)
+
+### Manage data
+
+- [How to manage Collection](./en/attu_collection.md)
+- [How to manage Data](./en/attu_data.md)
+- [How to manage Partition](./en/attu_partition.md)
+- [How to manage Query](./en/attu_query.md)
+- [How to manage Index](./en/attu_index.md)
+
+### Features
+
+- [Attu Overview](./en/attu_overview.md)
+- [vector search](./en/attu_search.md)
+- [System view](./en/attu_system.md)
+
+### FAQ
+
+- [FAQ](./en/attu_faq.md)

doc/en/attu.md

@@ -0,0 +1,49 @@
+---
+id: attu.md
+related_key: attu
+summary: Attu, an intuitive GUI for Milvus service management.
+---
+
+# Attu
+
+[Attu](https://github.com/zilliztech/attu) is an efficient open-source management tool for Milvus. It features an intuitive graphical user interface (GUI), allowing you to easily interact with your databases. With just a few clicks, you can visualize your cluster status, manage metadata, perform data queries, and much more.
+
+![Attu_overview](../assets/insight_overview.png)
+
+## Features
+
+Attu is under rapid development and new features are added on a weekly basis. We will release a new version every time when a new feature is ready.
+
+Below is the features we have to offer:
+
+- View milvus cluster statistics in a glance. (TBD)
+
+![view_cluster_statistics](../assets/view_cluster_statistics.png)
+
+- Browse, query, and manage collections in a simple and straightforward way.
+
+![manage_collections](../assets/manage_collections.png)
+
+- Perform CRUD or bulk operations with just a few clicks.
+
+![attu_operations](../assets/insight_operations.png)
+
+- Create vector index instantly.
+
+![attu_create_index](../assets/insight_create_index.png)
+
+- Conduct vector searches in a brand new way.
+
+![attu_conduct_search](../assets/insight_conduct_search.png)
+
+- New code-mode provides a better user experience for you.
+
+![code_mode](../assets/code_mode.png)
+
+Learn more about how to [install Attu](attu_install-docker.md).
+
+## Contribution
+
+Attu is an open-source project. All contributions are welcome. Pleae read our [Contribute guide](https://github.com/zilliztech/attu) before making contributions.
+
+If you find a bug or want to request a new feature, please create a [GitHub Issue](https://github.com/zilliztech/attu), and make sure that the same issue has not been created by someone else.

doc/en/attu_collection.md

@@ -0,0 +1,79 @@
+---
+id: attu_collection.md
+related_key: attu
+summary: Attu, an intuitive GUI for Milvus service management.
+---
+
+# Manage Collections with Attu
+
+This topic describes how to manage collections with Attu.
+
+## Create a collection
+
+1. Click the **Collection** icon on the left-side navigation pane and then click **Create Collection**. The **Create Collection** dialog box appears as shown below.
+
+![Create Collection dialog box](../assets/create_collection_dialog_box1.png)
+
+2. Enter the required information. This example creates a collection named `test` with a primary key field, a vector field, and a scalar field. You can add scalar fields as needed.
+
+![Create Collection dialog box](../assets/create_collection_dialog_box2.png)
+
+3. Click **Create** to create a collection.
+
+![Create Collection dialog box](../assets/create_collection_dialog_box3.png)
+
+## Delete a collection
+
+1. Tick the collection you want to delete in the data grid.
+2. Click the **Trash** icon and the **Delete Collection** dialog box appears as shown below.
+3. Type `delete` to confirm the deletion.
+4. Click **Delete** to delete the collection.
+
+<div class="alert caution">
+Deleting a collection is irreversible.
+</div>
+
+![Delete Collection dialog box](../assets/delete_collection.png)
+
+## Load a collection
+
+1. Hover on the collection you want to load, the **Load** icon appears on the right end of the entry.
+
+![Load Collection](../assets/load_collection1.png)
+
+2. Click the **Load** icon and the **Load Collection** dialog box appears.
+3. Click **Load** in the **Load Collection** dialog box.
+
+![Load Collection](../assets/load_collection2.png)
+
+4. Loading a collection might take a while. If successful, Loaded For Search appears in the Status column.
+
+![Load Collection](../assets/load_collection3.png)
+
+## Release a collection
+
+1. Hover on the loaded collection you want to release, the **Release** icon appears on the right end of the entry.
+
+![Release Collection](../assets/release_collection1.png)
+
+2. Click the **Release** icon and the **Release Collection** dialog box appears.
+3. Click **Release** in the **Release Collection** dialog box.
+4. If successful, the Status becomes **Unloaded** in the **Status** column.
+
+![Release Collection](../assets/release_collection2.png)
+
+## View the schema of a collection
+
+1. Click the name of the collection that you want to view the schema of, and the corresponding detail page appears.
+2. Click **Schema** on the detail page, which lists the information of all fields.
+
+Attributes of a schema include:
+
+- Field Name
+- Field Type
+- Dimension (Only applies to vector fields)
+- Index Type (Only applies to vector fields)
+- Index Parameters (Only applies to vector fields)
+- Collection description
+
+![Collection Schema](../assets/collection_schema.png)

doc/en/attu_data.md

@@ -0,0 +1,56 @@
+---
+id: attu_data.md
+related_key: attu
+summary: Attu, an intuitive GUI for Milvus service management.
+---
+
+# Manage Data with Attu
+
+This topic describes how to manage data with Attu.
+
+## Import data
+
+This example imports 20,000 rows of data. Importing data appends data instead of overwriting data.
+
+1. Click **Import Data** on the **Collection** page. The **Import Data** dialog box appears as shown below.
+
+![Import Data](../assets/insight_data1.png)
+
+2. Select the collection you want to import data to in the **Collection** dropdown list.
+3. Select the partition you want to import data to in the **Partition** dropdown list.
+4. Click **Choose CSV File** and choose a CSV file.
+
+<div class="alert note"> Ensure that the CSV file meets the following criteria:
+<ul>
+<li>Column names are the same as the field names specified in the schema;</li>
+<li>The file is smaller than 150MB and the row count is less than 100,000.</li>
+</ul>
+</div>
+
+5. After a legal CSV file is selected, you can then proceed by clicking **Next**.
+
+![Import Data](../assets/insight_data2.png)
+
+6. On the new dialog box, you can match the field names by clicking the corresponding cells in the dropdown lists.
+
+<div class="alert note">
+We recommend making the headers (column names) as the first row in your CSV file.
+</div>
+
+![Import Data](../assets/insight_data3.png)
+
+7. After confirming the column names corresponding to the field names, click **Import Data** to import the CSV file into Milvus. Importing data might take a while.
+
+![Import Data](../assets/insight_data4.png)
+
+8. If successful, the row count status updates in the Entity Count column of the collection. On the corresponding Partition tab page, the row count status updates in the Entity Count column of the partition your imported data in. It might take a while for the entity count to update.
+
+![Import Data](../assets/insight_data5.png)
+
+## Export Data
+
+1. Click **Data Query** on the **Collection** page. On the **Data Query** tab page, enter query conditions in the field and then click **Query** to retrieve all query results that match your query conditions.
+
+2. Click the **Download** icon to download the query results as a CSV file.
+
+![Export Data](../assets/insight_data6.png)

doc/en/attu_faq.md

@@ -0,0 +1,19 @@
+---
+id: attu_faq.md
+related_key: attu
+summary: Attu, an intuitive GUI for Milvus service management.
+---
+
+# Attu FAQ
+
+#### Why is Attu throwing a network error?
+
+A: Check whether you have assigned a correct value to `HOST_URL` in the `docker run` command. Alternatively, you can enter `{HOST_URL}/api/v1/healthy` in the address bar of your browser to check the network status of Attu.
+
+#### Why did Attu fail to connect to Milvus?
+
+A: Ensure that Milvus and Attu are on the same network.
+
+#### How do I use Attu with K8s?
+
+A: You can [install Attu while deploying Milvus with Helm](attu_install-helm.md).

doc/en/attu_index.md

@@ -0,0 +1,36 @@
+---
+id: attu_index.md
+related_key: attu
+summary: Attu, an intuitive GUI for Milvus service management.
+---
+
+# Manage Index with Attu
+
+This topic describes how to manage an index with Attu.
+
+## Create indexes
+
+This example builds an IVF_FLAT index with Euclidean distance as the similarity metrics and an `nlist` value of `1024`.
+
+1. Click **Schema** on the **Collection** page. On the **Schema** tab page, click **CREATE INDEX** and the **Create Index** dialog box appears.
+
+2. In the **Create Index** dialog box, select **IVF_FLAT** from the **Index Type** dropdown list, select **L2** from the **Metric Type** dropdown list, and enter `1024` in the `nlist` field.
+
+3. (Optional) Turn on **View Code** and the **Code View** page appears. You can check the code in Python or Node.js as you want.
+
+4. Click **Create** to create the index.
+
+If successful, the type of the index you created appears in the **Index Type** column for the vector field.
+
+![Create Index](../assets/insight_index1.png)
+
+![Create Index](../assets/insight_index2.png)
+
+## Delete indexes
+
+1. Click the **Trash** icon in the **Index Type** column and the **Delete Index** dialog box appears.
+2. Enter `delete` to confirm the deletion and click **Delete** to delete the indexes.
+
+If successful, **CREATE INDEX** button appears in the Index Type column.
+
+![Delete Index](../assets/insight_index3.png)

doc/en/attu_install-docker.md

@@ -0,0 +1,39 @@
+---
+id: attu_install-docker.md
+label: Install with Docker Compose
+order: 0
+group: attu
+related_key: attu
+summary: Learn how to install Attu with Docker Compose to manage your Milvus service.
+---
+
+# Install Attu
+
+This topic describes how to install Attu, an efficient open-source management tool for Milvus.
+
+{{tab}}
+
+## Prerequisites
+
+- Milvus installed on [your local device](install_standalone-docker.md) or [cluster](install_cluster-docker.md).
+- Docker 19.03 or later
+
+<div class="alert note">
+Attu only supports Milvus 2.x.
+</div>
+
+## Start a Attu instance
+
+```Apache
+docker run -p 8000:3000 -e HOST_URL=http://{ your machine IP }:8000 -e MILVUS_URL={your machine IP}:19530 zilliz/attu:latest
+```
+
+Once you start the docker, visit `http://{ your machine IP }:8000` in your browser, and click **Connect** to enter the Attu service.
+
+![Attu_install](../assets/insight_install.png)
+
+## Contribution
+
+Attu is an open-source project. All contributions are welcome. Please read our [Contribute guide](https://github.com/zilliztech/attu) before making contributions.
+
+If you find a bug or want to request a new feature, please create a [GitHub Issue](https://github.com/zilliztech/attu), and make sure that the same issue has not been created by someone else.

doc/en/attu_install-helm.md

@@ -0,0 +1,162 @@
+---
+id: attu_install-helm.md
+label: Install with Helm Chart
+order: 1
+group: attu
+related_key: attu
+summary: Learn how to install Attu with Helm Chart to manage your Milvus service.
+---
+
+# Install Attu
+
+This topic describes how to install Attu, an efficient open-source management tool for Milvus.
+
+{{tab}}
+
+## Prerequisites
+
+- Kubernetes 1.16 or later
+- Helm 3.0.0 or later
+
+<div class="alert note">
+Attu only supports Milvus 2.x.
+</div>
+
+## Install Helm Chart for Milvus
+
+Helm is a Kubernetes package manager that can help you deploy Milvus quickly.
+
+1. Add Milvus Helm repository.
+
+```
+$ helm repo add milvus https://milvus-io.github.io/milvus-helm/
+```
+
+2. Update charts locally.
+
+```
+$ helm repo update
+```
+
+## Install Attu while installing Milvus
+
+Start Milvus and Attu with Helm by specifying the release name, the chart, and the parameters that indicate the installation and service mode of Attu. This topic uses `my-release` as the release name. To use a different release name, replace `my-release` in the command.
+
+Attu provides services in the following three modes among which you can select to suit your scenario. It is recommended to use Ingress mode. Port-forward mode is suggested to be used in test environment only.
+
+- [Ingress mode](#Ingress-mode)
+- [LoadBalancer mode](#LoadBalancer-mode)
+- [Port-forward mode](#Port-forward-mode)
+
+### Ingress mode
+
+Ensure that you have integrated Ingress controller in your Kubernetes cluster.
+
+1. Install Milvus and Attu.
+
+```
+helm install my-release milvus/milvus --set insight.enabled=true
+```
+
+2. Check the established Ingress.
+
+```
+kubectl get ingress
+```
+
+3. Check the addresses that correspond to `my-release-milvus-insight` in the returned result.
+
+```
+NAME                          CLASS    HOSTS                  ADDRESS                               PORTS   AGE
+my-release-milvus-insight    <none>   milvus-insight.local   10.100.32.1,10.100.32.2,10.100.32.3   80      22h
+```
+
+4. Configure DNS on the device that requires the Attu service by mapping the path `milvus-insight.local` onto any of the addresses returned above in the system file `/etc/hosts`.
+
+```
+10.100.32.1     milvus-insight.local
+```
+
+5. Visit `http://milvus-insight.local` in your browser, and click **Connect** to enter the Attu service.
+
+![Attu_install](../assets/insight_install.png)
+
+### LoadBalancer mode
+
+Ensure that you have integrated LoadBalancer in your Kubernetes cluster.
+
+1. Install Milvus and Attu.
+
+```
+helm install my-release milvus/milvus --set insight.enabled=true --set insight.service.type=LoadBalancer --set insight.ingress.enabled=false
+```
+
+2. Check the Attu service.
+
+```
+kubectl get svc
+```
+
+3. Check the external IP of the service `my-release-milvus-insight` in the returned results.
+
+```
+NAME                                    TYPE           CLUSTER-IP      EXTERNAL-IP   PORT(S)
+my-release-etcd                        ClusterIP      10.96.106.84    <none>        2379/TCP,2380/TCP                     117s
+my-release-etcd-headless               ClusterIP      None            <none>        2379/TCP,2380/TCP                     117s
+my-release-milvus                      ClusterIP      10.96.230.238   <none>        19530/TCP,9091/TCP                    117s
+my-release-milvus-datacoord            ClusterIP      10.96.75.27     <none>        13333/TCP,9091/TCP                    117s
+my-release-milvus-datanode             ClusterIP      None            <none>        9091/TCP                              117s
+my-release-milvus-indexcoord           ClusterIP      10.96.183.151   <none>        31000/TCP,9091/TCP                    117s
+my-release-milvus-indexnode            ClusterIP      None            <none>        9091/TCP                              117s
+my-release-milvus-insight              LoadBalancer   10.96.79.103    10.98.0.16    3000:30413/TCP                        117s
+my-release-milvus-querycoord           ClusterIP      10.96.204.140   <none>        19531/TCP,9091/TCP                    117s
+my-release-milvus-querynode            ClusterIP      None            <none>        9091/TCP                              117s
+my-release-milvus-rootcoord            ClusterIP      10.96.142.19    <none>        53100/TCP,9091/TCP                    117s
+my-release-minio                       ClusterIP      10.96.55.66     <none>        9000/TCP                              117s
+my-release-minio-svc                   ClusterIP      None            <none>        9000/TCP                              117s
+my-release-pulsar-bookkeeper           ClusterIP      None            <none>        3181/TCP                              117s
+my-release-pulsar-broker               ClusterIP      10.96.177.151   <none>        8080/TCP,6650/TCP,8443/TCP,6651/TCP   117s
+my-release-pulsar-proxy                ClusterIP      10.96.148.241   <none>        8080/TCP,6650/TCP,8000/TCP            117s
+my-release-pulsar-zookeeper            ClusterIP      None            <none>        2888/TCP,3888/TCP,2181/TCP            117s
+my-release-pulsar-zookeeper-ca         ClusterIP      10.96.100.254   <none>        2888/TCP,3888/TCP,2181/TCP            117s
+```
+
+4. Configure DNS on the device that requires the Attu service by mapping the path `my-release-milvus-insight` onto its external IP returned above in the system file `/etc/hosts`.
+
+```
+10.98.0.16 my-release-milvus-insight
+```
+
+5. Visit `http://my-release-milvus-insight:3000/connect` in your browser, and click **Connect** to enter the Attu service.
+
+![Attu_install](../assets/insight_install.png)
+
+### Port-forward mode
+
+1. Install Milvus and Attu.
+
+```
+helm install my-release milvus/milvus --set insight.enabled=true  --set insight.ingress.enabled=false
+```
+
+2. Forward the Attu service to local port `3000`.
+
+```
+kubectl port-forward service/my-release-milvus-insight 3000
+```
+
+3. Configure DNS on the device that forwards the Attu service by mapping the path `my-release-milvus-insight` onto `127.0.0.1` in the system file `/etc/hosts`.
+
+```
+127.0.01 my-release-milvus-insight
+```
+
+4. Visit `http://my-release-milvus-insight:3000/connect` in your browser, and click **Connect** to enter the Attu service.
+
+![Attu_install](../assets/insight_install.png)
+
+## Contribution
+
+Attu is an open-source project. All contributions are welcome. Please read our [Contribute guide](https://github.com/zilliztech/attu) before making contributions.
+
+If you find a bug or want to request a new feature, please create a [GitHub Issue](https://github.com/zilliztech/attu), and make sure that the same issue has not been created by someone else.

doc/en/attu_overview.md

@@ -0,0 +1,20 @@
+---
+id: attu_overview.md
+related_key: attu
+summary: Attu, an intuitive GUI for Milvus service management.
+---
+
+# Attu Overview Page
+
+This topic describes the Home page of Attu.
+
+Attu consists of **Overview** page, **Collection** page, **Vector Search** page, and **System View** page, corresponding to the four icons on the left-side navigation pane respectively.
+
+The Overview page lists the following information:
+
+1. Loaded Collections: the number of loaded collections.
+2. All Collections: the total number of collections.
+3. Data: the total number of entities.
+4. **Loaded For Search** card: an interactive shortcut panel that allows you to perform a [vector search](attu_search.md) on or [release the collection](attu_collection.md).
+
+![Attu Overview](../assets/insight_overview_page.png)

doc/en/attu_partition.md

@@ -0,0 +1,35 @@
+---
+id: attu_partition.md
+related_key: attu
+summary: Attu, an intuitive GUI for Milvus service management.
+---
+
+# Manage Partitions with Attu
+
+This topic describes how to manage partitions with Attu.
+
+Milvus creates a partition automatically after a collection is created, which cannot be deleted.
+
+## Create a partition
+
+1. Click **Partitions** tab on the **Collection** page.
+2. Click **Create Partition** on the **Partitions** tab page, and the **Create Partition** dialog box appears as shown below.
+3. In the **Create Partition** dialog box, enter the new partition name in the **Name** field.
+4. Click **Create** to create a partition.
+
+![Create Partition](../assets/insight_partition1.png)
+
+If successful, the new partition appears on the **Partitions** tab page.
+
+![Create Partition](../assets/insight_partition2.png)
+
+Choose the Default partition or the newly created partition to store imported data as needed.
+
+## Delete a partition
+
+1. Tick the partition you want to delete.
+2. Click the **Trash** icon and the **Delete Partition** dialog box appears as shown below.
+3. Type `delete` to confirm the deletion.
+4. Click **Delete** to delete the partition.
+
+![Delete Partition](../assets/insight_partition3.png)

doc/en/attu_query.md

@@ -0,0 +1,37 @@
+---
+id: attu_query.md
+related_key: attu
+summary: Attu, an intuitive GUI for Milvus service management.
+---
+
+# Query Data with Attu
+
+This topic describes how to query data with Attu.
+
+## Query data with advanced filters
+
+1. Click the entry of the collection that you want to query data in, and the corresponding detail page appears.
+2. On the **Data Query** tab page, click the **Filter** icon and the **Advanced Filter** dialog box appears.
+3. Specify a complicated query condition such as **color > 10 && color < 20** by using the **Field Name** dropdown lists, the **Logic** dropdown lists, the **Value** fields, and the **AND** operator. Then click **Apply** Filter to apply the query condition.
+
+![Query Data](../assets/insight_query1.png)
+
+3. Click **Query** to retrieve all query results that match the query condition.
+
+![Query Data](../assets/insight_query2.png)
+
+## Query data with Time Travel
+
+TBD (Not supported yet)
+
+## Delete data
+
+1. Tick the entities you want to delete and click the **Trash** icon.
+2. Type `delete` to confirm the deletion in the **Delete entities** dialog box.
+3. Click **Delete** to delete the selected entities.
+
+![Delete Data](../assets/insight_query3.png)
+
+You can perform a query to retrieve the deleted entities. No results will be returned if the deletion is successful.
+
+![Delete Data](../assets/insight_query4.png)

doc/en/attu_search.md

@@ -0,0 +1,47 @@
+---
+id: attu_search.md
+related_key: attu
+summary: Attu, an intuitive GUI for Milvus service management.
+---
+
+# Search Data with Attu
+
+This topic describes how to search data with Attu.
+
+## Conduct a vector similarity search
+
+On the basis of the regular vector similarity search, you can perform hybrid search of search with Time Travel.
+
+### Load the collection to memory
+
+All CRUD operations within Milvus are executed in memory. Load the collection to memory before conducting a vector similarity search. See [Load a collection](attu_collection.md#Load-a-collection) for more instruction.
+
+![Search Data](../assets/insight_search1.png)
+
+### Set search parameters
+
+1. Select the collection and the vector field you wish to search in in the dropdown lists of the **Choose collection and field** section.
+2. In the **Enter vector value** field, enter a vector (or vectors) with the same dimensions of the selected field as the target vector(s) to search with.
+3. In the **Set search parameters** section, specify the specific parameter(s) to the index and other search-related parameters.
+
+![Search Data](../assets/insight_search2.png)
+
+### Hybrid search with advanced filters (optional)
+
+Click **Advanced Filter** and the **Advanced Filter** dialog box appears. You can use the **AND** or **OR** operators to combine multiple conditions into a compound condition. The filter expression updates automatically with any changes to the conditions. See [boolean expression rule](boolean.md) for more information.
+
+![Search Data](../assets/insight_search3.png)
+
+### Search with Time Travel (optional)
+
+Milvus maintains a timeline for all data insert and delete operations. It allows users to specify a timestamp in a search to retrieve a data view at a specified point in time.
+
+1. Click **Time Travel**, and select a time point in the dialog box that appears.
+
+![Search Data](../assets/insight_search4.png)
+
+2. Specify the number of search results to return in the **TopK** dropdown list.
+3. Click **Search** to retrieve the nearest search results, which indicate the most similar vectors.
+
+![Search Data](../assets/insight_search5.png)
+![Search Data](../assets/insight_search6.png)

doc/en/attu_system.md

@@ -0,0 +1,38 @@
+---
+id: attu_system.md
+related_key: attu
+summary: Attu, an intuitive GUI for Milvus service management.
+---
+
+# Monitor System with Attu
+
+This topic describes how to monitor your Milvus system with Attu.
+
+## System view
+
+Click the **System View** icon on the left-side navigation pane to visit the System View page.
+
+![System View](../assets/insight_system1.png)
+
+The System View dashboard consists of the following tiles:
+
+- **Disk**: Shows the usage of storage space.
+- **Memory**: Shows the usage of memory.
+- **Qps**: Shows the last 10 monitored QPS. Each node represents a time point. Hover on a node to view the QPS monitored at the time point.
+- **Latency**: Shows the last 10 monitored latency. Each node represents a time point. Hover on a node to view the latency monitored at the time point.
+
+![System View](../assets/insight_system2.png)
+![System View](../assets/insight_system3.png)
+
+- **Topology**: Shows the structure of your Milvus instance. Click Milvus node or a coordinator node and the corresponding information appears in the Info tile on the right.
+- **Info**: Shows the hardware, system, and configuration information of the selected node.
+
+## Node list view
+
+All nodes managed by their parent coordinator node are listed in the table. You can sort the nodes by metrics including **CPU Core Count**, **CPU Core Usage**, **Disk Usage**, and **Memory Usage**.
+
+![Node List View](../assets/insight_system4.png)
+
+Right to the table is a mini topology showing the selected node and its coordinator node. Under the mini topology is a tile showing relevant information.
+
+Click the down arrow to collapse rows in the table.

doc/zh-CN/attu.md

@@ -0,0 +1,49 @@
+---
+id: attu.md
+related_key: attu
+summary: Attu, an intuitive GUI for Milvus service management.
+---
+
+# Attu
+
+[Attu](https://github.com/zilliztech/attu) 是 Milvus 的高效开源管理工具. 它具有直观的图形用户界面 (GUI)，可让您轻松与数据库进行交互。 只需点击几下，您就可以可视化集群状态、管理元数据、执行数据查询等等。
+
+![Attu_overview](../assets/insight_overview.png)
+
+## 特点
+
+Attu 正在快速开发中，每周都会添加新功能。每当新功能准备就绪时，我们都会发布新版本。
+
+以下是我们必须提供的功能：
+
+- Milvus 集群统计信息一目了然。
+
+![view_cluster_statistics](../assets/view_cluster_statistics.png)
+
+- 以简单直接的方式浏览、查询和管理集合。
+
+![manage_collections](../assets/manage_collections.png)
+
+- 只需单击几下即可执行 CRUD 或批量操作。
+
+![attu_operations](../assets/insight_operations.png)
+
+- 创建矢量索引。
+
+![attu_create_index](../assets/insight_create_index.png)
+
+- 以全新的方式进行向量相似度搜索。
+
+![attu_conduct_search](../assets/insight_conduct_search.png)
+
+- 新的代码模式为您提供更好的用户体验。
+
+![code_mode](../assets/code_mode.png)
+
+阅读[安装 Attu](attu_install-docker.md) 以了解更多。
+
+## 贡献
+
+Attu 是一个开源项目。欢迎所有贡献者。在提交代码前，请阅读我们的[贡献指南](https://github.com/zilliztech/attu)。
+
+如果您发现错误或想要请求新功能， 请创建一个 [GitHub Issue](https://github.com/zilliztech/attu)，并确保其他人没有创建相同的问题。

doc/zh-CN/attu_collection.md

@@ -0,0 +1,79 @@
+---
+id: attu_collection.md
+related_key: attu
+summary: Attu, an intuitive GUI for Milvus service management.
+---
+
+# 使用 Attu 管理 Collections
+
+这片文章将会描述如何使用 Attu 管理 collections。
+
+## 创建 collection
+
+1. 点击左侧导航的 **Collection** 标签，然后点击 **Create Collection**。如下图所示，将会出现 **Create Collection** 对话框。
+
+![Create Collection dialog box](../assets/create_collection_dialog_box1.png)
+
+2. 填写相应表单，这个例子将创建名为 test 的 collection，这个 collection 有一个主键字段，一个向量字段，一个标量字段。你也可以继续添加需要的标量字段。
+
+![Create Collection dialog box](../assets/create_collection_dialog_box2.png)
+
+3. 点击 **Create** 就可以创建我们第一个 collection。
+
+![Create Collection dialog box](../assets/create_collection_dialog_box3.png)
+
+## 删除 collection
+
+1. 在表格里，选中需要删除的 collection。
+2. 点击 **Trash** 图标，如下图所示，将会出现 **Delete Collection** 对话框。
+3. 输入 `delete` 以确认删除操作。
+4. 点击 **Delete** 将会删除选中的所有 collection。
+
+<div class="alert caution">
+删除操作是不可回滚的。
+</div>
+
+![Delete Collection dialog box](../assets/delete_collection.png)
+
+## 加载 collection
+
+1. 鼠标悬浮在需要加载的 collection 行上，**Load** 图标将会出现在该行末尾。
+
+![Load Collection](../assets/load_collection1.png)
+
+2. 点击 **Load** 图标，将会出现 **Load Collection** 对话框。
+3. 点击对话框中的 **Load** 按钮。
+
+![Load Collection](../assets/load_collection2.png)
+
+4. 加载 collection 数据到缓存中可能需要一些时间。当加载完成后，**Status** 列将会显示 **Loaded For Search**。
+
+![Load Collection](../assets/load_collection3.png)
+
+## 释放 collection
+
+1. 鼠标悬浮在需要加载的 collection 行上，**Release** 图标将会出现在该行末尾。
+
+![Release Collection](../assets/release_collection1.png)
+
+2. 点击 **Release** 图标，将会出现 **Release Collection** 对话框。
+3. 点击对话框中的 **Release** 按钮以释放 collection。
+4. 如果释放成功，**Status** 列将显示 **Unloaded**。
+
+![Release Collection](../assets/release_collection2.png)
+
+## 查看 collection 的 schema
+
+1. 点击数据行的 collection name，跳转到 collection 详情页面。
+2. 在详情页点击 **Schema**，就可以看到 schema 的相关信息。
+
+schema 的属性包含：
+
+- Field Name
+- Field Type
+- Dimension (适用于向量字段)
+- Index Type (适用于向量字段)
+- Index Parameters (适用于向量字段)
+- Description
+
+![Collection Schema](../assets/collection_schema.png)

doc/zh-CN/attu_data.md

@@ -0,0 +1,61 @@
+---
+id: attu_data.md
+related_key: attu
+summary: Attu, an intuitive GUI for Milvus service management.
+---
+
+# 使用 Attu 管理数据
+
+这片文章将会描述如何使用 Attu 管理数据。
+
+## 导入数据
+
+这个例子将导入 20,000 行数据。导入数据是追加数据，而不是覆盖数据。
+
+1. 在 **Collection** 页面，点击 **Import Data** 按钮。如下图所示，将会出现 **Import Data** 对话框。
+
+![Import Data](../assets/insight_data1.png)
+
+2. 在 **Collection** 下拉列表框中，选择需要的 collection 来导入数据。
+3. 在 **Partition** 下拉列表框中，选择需要的 partition 来导入数据。
+4. 点击 **Choose CSV File** 选择 CSV 文件。
+
+<div class="alert note">
+ 需要保证 CSV 文件符合以下条件：
+<ul>
+<li>
+列名与 collection 中 schema 的字段名完全一致。
+</li>
+<li>
+上传的 CSV 文件需要小于 150MB，以及行的总数小于 100,000。
+</li>
+</ul>
+</div>
+
+5. 上传了一个合法的 CSV 文件以后，便可以点击 **Next**。
+
+![Import Data](../assets/insight_data2.png)
+
+6. 在新的对话框中，你可以选择在下拉列表框中选择列名。
+
+<div class="alert note">
+我们推荐在 CSV 文件的首行设置你的列名。
+</div>
+
+![Import Data](../assets/insight_data3.png)
+
+7. 在确认完列名与 collection 的字段一致后，点击 **Import Data** 将会导入 CSV 文件中的数据至 Milvus。 导入数据可能需要一段时间。
+
+![Import Data](../assets/insight_data4.png)
+
+8. 导入成功后，collection 的 Entity Count 列将会更新为导入数据的总数。同时在 Partition 页面，导入数据时选择的 partition 的 Entity Count 列也会相应的更新。 这个更新可能会需要一些时间。
+
+![Import Data](../assets/insight_data5.png)
+
+## 导出数据
+
+1. 在 **Collection** 页面， 点击 **Data Query** 页签。进入 **Data Query** 页面， 输入你需要的 query 条件，然后点击 **Query** 就能得到符合条件的所有数据。
+
+2. 点击 **Download** 图标，将会下载这些数据, 作为 CSV 文件保存到本地。
+
+![Export Data](../assets/insight_data6.png)

doc/zh-CN/attu_faq.md

@@ -0,0 +1,19 @@
+---
+id: attu_faq.md
+related_key: attu
+summary: Attu, an intuitive GUI for Milvus service management.
+---
+
+# Attu FAQ
+
+#### Attu 为什么报网络错误?
+
+答：请确认在执行 `docker run` 命令时，传入了正确的 `HOST_URL` 值。你也可以在浏览器输入 `{HOST_URL}/api/v1/healthy` 来确认 Attu 的服务状态。
+
+#### 为什么 Attu 连接不上 Milvus？
+
+答：请确保 Milvus 和 Attu 在同一网络。
+
+#### 我该如何在 k8s 中使用 Attu？
+
+答：请参考[使用 Helm Chart 安装 Attu](https://milvus.io/cn/docs/v2.0.0/attu_install-helm.md)。

doc/zh-CN/attu_index.md

@@ -0,0 +1,36 @@
+---
+id: attu_index.md
+related_key: attu
+summary: Attu, an intuitive GUI for Milvus service management.
+---
+
+# 使用 Attu 管理索引
+
+这片文章将会描述如何使用 Attu 管理索引。
+
+## 创建索引
+
+这个例子将会创建 IVF_FLAT 索引，使用 Euclidean 距离作为相似性度量，以及 `nlist` 值为 `1024`。
+
+1. 在 **Collection** 页面，点击 **Schema** 页签。 然后在 **Schema** 页面，点击 **CREATE INDEX**，将会出现 **Create Index** 对话框。
+
+2. 在 **Create Index** 对话框中，在 **Index Type** 下拉列表框中，选择 **IVF_FLAT**， 在 **Metric Type** 下拉列表框中，选择 **L2**，最后在 `nlist` 输入框中输入 `1024`。
+
+3. 你可以点击 **View Code** 开关，将会看到 **Code View** 页面。你可以在这个页面中看到相应的 Python 或者 Node.js 代码。
+
+4. 点击 **Create** 创建索引。
+
+如果创建成功，你创建的索引类型将会出现在向量字段的 **Index Type** 列中。
+
+![Create Index](../assets/insight_index1.png)
+
+![Create Index](../assets/insight_index2.png)
+
+## 删除索引
+
+1. 点击 **Index Type** 列中的 **Trash** 图标，将会出现 **Delete Index** 对话框。
+2. 输入 `delete` 来确认删除操作，然后点击 **Delete** 来删除索引。
+
+如果删除成功，**CREATE INDEX** 按钮将会出现在 **Index Type** 列中。
+
+![Delete Index](../assets/insight_index3.png)

doc/zh-CN/attu_install-docker.md

@@ -0,0 +1,40 @@
+---
+id: attu_install-docker.md
+label: Install with Docker Compose
+order: 0
+group: attu
+related_key: attu
+summary: Learn how to install Attu with Docker Compose to manage your Milvus service.
+---
+
+# 安装 Attu
+
+这篇文章将描述如何安装 Attu - 一个开源的 MIlvus 管理工具。
+
+{{tab}}
+
+## 先决条件
+
+- 已安装 Milvus [单机版](install_standalone-docker.md) 或者 [分布式版](install_cluster-docker.md)。
+- Docker 版本 19.03 或者更新的版本。
+
+<div class="alert note">
+Attu 只支持 Milvus 2.x。
+</div>
+
+## 运行 Attu
+
+```Apache
+docker run -p 8000:3000 -e HOST_URL=http://{ your machine IP }:8000 -e MILVUS_URL={your machine IP}:19530 zilliz/attu:latest
+```
+
+一旦你成功运行了 Attu docker, 在浏览器输入 `http://{ your machine IP }:8000`，
+然后点击 **Connect** 按钮连接 Milvus。
+
+![Attu_install](../assets/insight_install.png)
+
+## 欢迎成为贡献者
+
+Attu 是一个开源项目。所有人都欢迎成为贡献者。在提交代码前，可以参考[贡献导读](https://github.com/zilliztech/attu)。
+
+如果你发现一个 bug 或者想添加新功能，请创建一个 [GitHub Issue](https://github.com/zilliztech/attu)，同时请确认仓库中不存在类似的 issue。

doc/zh-CN/attu_install-helm.md

@@ -0,0 +1,162 @@
+---
+id: attu_install-helm.md
+label: Install with Helm Chart
+order: 1
+group: attu
+related_key: attu
+summary: Learn how to install Attu with Helm Chart to manage your Milvus service.
+---
+
+# 安装 Attu
+
+本文介绍如何安装 Attu 工具。Attu 是 Milvus 的高效开源管理工具。
+
+{{tab}}
+
+## 先决条件
+
+- Kubernetes 1.16 或更高版本
+- Helm 3.0.0 或更高版本
+
+<div class="alert note">
+Attu 只支持 Milvus 2.x。
+</div>
+
+## 为 Milvus 安装 Helm Chart
+
+Helm 是一个 Kubernetes 包管理器，可以帮助您快速部署 Milvus。
+
+1. 添加 Milvus Helm 存储库。
+
+```
+$ helm repo add milvus https://milvus-io.github.io/milvus-helm/
+```
+
+2. 在本地更新图表。
+
+```
+$ helm repo update
+```
+
+## 在安装 Milvus 时安装 Attu
+
+通过指定版本名称、图表和参数来启动 Milvus 和 Attu，这些参数表明 Attu 的安装和服务模式。本主题使用 `my-release` 作为发布名称。要使用不同的发布名称，请在命令中替换 `my-release`。
+
+Attu 提供以下三种服务模式，您可以根据自己的场景选择其中一种。建议使用 Ingress 模式。端口转发模式建议仅在测试环境中使用。
+
+- [Ingress 模式](#Ingress-mode)
+- [LoadBalancer 模式](#LoadBalancer-mode)
+- [Port-forward 模式](#Port-forward-mode)
+
+### Ingress 模式
+
+确保在 Kubernetes 集群中集成了 Ingress 控制器。
+
+1. 安装 Milvus 和 Attu。
+
+```
+helm install my-release milvus/milvus --set insight.enabled=true
+```
+
+2. 检查已建立的 Ingress。
+
+```
+kubectl get ingress
+```
+
+3. 检查返回结果中对应于 `my-release-milvus-insight` 的地址。
+
+```
+NAME                          CLASS    HOSTS                  ADDRESS                               PORTS   AGE
+my-release-milvus-insight    <none>   milvus-insight.local   10.100.32.1,10.100.32.2,10.100.32.3   80      22h
+```
+
+4. 通过将路径 `milvus-insight.local` 映射到系统文件 `/etc/hosts` 中返回的任何地址，在需要 Attu 服务的设备上配置 DNS。
+
+```
+10.100.32.1     milvus-insight.local
+```
+
+5. 访问 `http://milvus-insight.local`，然后单击 **Connect** 进入 Attu 服务。
+
+![Attu_install](../assets/insight_install.png)
+
+### LoadBalancer 模式
+
+确保在 Kubernetes 集群中集成了 LoadBalancer。
+
+1. 安装 Milvus and Attu。
+
+```
+helm install my-release milvus/milvus --set insight.enabled=true --set insight.service.type=LoadBalancer --set insight.ingress.enabled=false
+```
+
+2. 检查 Attu 服务。
+
+```
+kubectl get svc
+```
+
+3. 在返回的结果中检查服务 `my-release-milvus-insight` 的外部 IP。
+
+```
+NAME                                    TYPE           CLUSTER-IP      EXTERNAL-IP   PORT(S)
+my-release-etcd                        ClusterIP      10.96.106.84    <none>        2379/TCP,2380/TCP                     117s
+my-release-etcd-headless               ClusterIP      None            <none>        2379/TCP,2380/TCP                     117s
+my-release-milvus                      ClusterIP      10.96.230.238   <none>        19530/TCP,9091/TCP                    117s
+my-release-milvus-datacoord            ClusterIP      10.96.75.27     <none>        13333/TCP,9091/TCP                    117s
+my-release-milvus-datanode             ClusterIP      None            <none>        9091/TCP                              117s
+my-release-milvus-indexcoord           ClusterIP      10.96.183.151   <none>        31000/TCP,9091/TCP                    117s
+my-release-milvus-indexnode            ClusterIP      None            <none>        9091/TCP                              117s
+my-release-milvus-insight              LoadBalancer   10.96.79.103    10.98.0.16    3000:30413/TCP                        117s
+my-release-milvus-querycoord           ClusterIP      10.96.204.140   <none>        19531/TCP,9091/TCP                    117s
+my-release-milvus-querynode            ClusterIP      None            <none>        9091/TCP                              117s
+my-release-milvus-rootcoord            ClusterIP      10.96.142.19    <none>        53100/TCP,9091/TCP                    117s
+my-release-minio                       ClusterIP      10.96.55.66     <none>        9000/TCP                              117s
+my-release-minio-svc                   ClusterIP      None            <none>        9000/TCP                              117s
+my-release-pulsar-bookkeeper           ClusterIP      None            <none>        3181/TCP                              117s
+my-release-pulsar-broker               ClusterIP      10.96.177.151   <none>        8080/TCP,6650/TCP,8443/TCP,6651/TCP   117s
+my-release-pulsar-proxy                ClusterIP      10.96.148.241   <none>        8080/TCP,6650/TCP,8000/TCP            117s
+my-release-pulsar-zookeeper            ClusterIP      None            <none>        2888/TCP,3888/TCP,2181/TCP            117s
+my-release-pulsar-zookeeper-ca         ClusterIP      10.96.100.254   <none>        2888/TCP,3888/TCP,2181/TCP            117s
+```
+
+4. 通过将路径 `my-release-milvus-insight` 映射到系统文件 `/etc/hosts` 中返回的外部 IP 上，在需要 Attu 服务的设备上配置 DNS。
+
+```
+10.98.0.16 my-release-milvus-insight
+```
+
+5. 在您的浏览器中访问 `http://my-release-milvus-insight:3000/connect`，然后单击**连接**进入 Attu 服务。
+
+![Attu_install](../assets/insight_install.png)
+
+### Port-forward 模式
+
+1. 安装 Milvus 和 Attu。
+
+```
+helm install my-release milvus/milvus --set insight.enabled=true  --set insight.ingress.enabled=false
+```
+
+2. 将 Attu 服务转发到本地端口 `3000`。
+
+```
+kubectl port-forward service/my-release-milvus-insight 3000
+```
+
+3. 在转发 Attu 服务的设备上配置 DNS，将路径 `my-release-milvus-insight` 映射到系统文件 `/etc/hosts` 中的 `127.0.0.1`。
+
+```
+127.0.01 my-release-milvus-insight
+```
+
+4. 在您的浏览器中访问 `http://my-release-milvus-insight:3000/connect`，然后单击**连接**进入 Attu 服务。
+
+![Attu_install](../assets/insight_install.png)
+
+## 贡献
+
+Attu 是一个开源项目。欢迎所有贡献。投稿前请先阅读我们的[投稿指南](https://github.com/zilliztech/attu)。
+
+如果你发现一个 bug 或者想请求一个新特性，请创建一个 [GitHub Issue](https://github.com/zilliztech/attu)，并确保相同的问题没有由其他人创建。

doc/zh-CN/attu_overview.md

@@ -0,0 +1,20 @@
+---
+id: attu_overview.md
+related_key: attu
+summary: Attu, an intuitive GUI for Milvus service management.
+---
+
+# Attu Overview 页面
+
+这个篇文章将描述 Attu 的首页。
+
+Attu 主要有 **Overview** 页面、**Collection** 页面、**Vector Search** 页面、和 **System View** 页面，对应左侧导航栏的四个图标。
+
+**Overview** 页面主要列出了以下信息：
+
+1. 所有加载到内存的 collection 数量
+2. 所有 collection 数量
+3. 所有导入到 Milvus 的数据
+4. **Loaded For Search** 卡片：一个交互式快捷面板，允许你执行[向量搜索](attu_search.md)或者[释放 Collection](attu_collection.md)
+
+![Attu Overview](../assets/insight_overview_page.png)

doc/zh-CN/attu_partition.md

@@ -0,0 +1,35 @@
+---
+id: attu_partition.md
+related_key: attu
+summary: Attu, an intuitive GUI for Milvus service management.
+---
+
+# 使用 Attu 管理 Partitions
+
+这个篇文章将描述 Attu 如何管理 partition。
+
+Milvus 会在创建 collection 成功以后，自动创建一个默认的，不可删除的 partition。
+
+## 创建 partition
+
+1. 在 **Collection** 详情页面，点击 **Partitions** 页签。
+2. 在 **Partitions** 页签，点击 **Create Partition** ，如下图所示，将会出现 **Create Partition** 对话框。
+3. 在 **Create Partition** 对话框中，在 **Name** 输入框内，输入新的 partition 名称。
+4. 点击 **Create** 按钮，创建一个 partition。
+
+![Create Partition](../assets/insight_partition1.png)
+
+如果创建成功，新的 partition 会出现在 **Partitions** 页面中。
+
+![Create Partition](../assets/insight_partition2.png)
+
+这样就可以选择任意一个 partition 来存储数据。
+
+## 删除 partition
+
+1. 选择需要删除的 partition。
+2. 点击 **Trash** 图标，如下图所示，将会出现 **Delete Partition** 对话框。
+3. 输入`delete` 确认删除操作。
+4. 点击 **Delete** 删除 partition。
+
+![Delete Partition](../assets/insight_partition3.png)

doc/zh-CN/attu_query.md

@@ -0,0 +1,33 @@
+---
+id: attu_query.md
+related_key: attu
+summary: Attu, an intuitive GUI for Milvus service management.
+---
+
+# 使用 Attu 获取数据
+
+这个篇文章将描述 Attu 如何获取数据。
+
+## 获取数据
+
+1. 点击需要获取数据的 collection，就可以跳转到相应到详情页面。
+2. 在 **Data Query** 页面，点击 **Filter** 标签，将会出现 **Advanced Filter** 对话框。
+3. 通过表单，选择你需要的筛选条件，例如 **color > 10 && color < 20**。 点击 **Apply** 确认筛选条件。
+
+![Query Data](../assets/insight_query1.png)
+
+1. 点击 **Query**，获取所有符合条件的数据。
+
+![Query Data](../assets/insight_query2.png)
+
+## 删除数据
+
+1. 选择需要删除的数据，然后点击 **Trash** 图标。
+2. 在 **Delete entities** 对话框中，输入 `delete` 确认删除操作。
+3. 点击 **Delete** 删除选中的数据。
+
+![Delete Data](../assets/insight_query3.png)
+
+删除成功后，通过 **Query** 获取已删除的数据，结果将不会包含已删除的数据。
+
+![Delete Data](../assets/insight_query4.png)

doc/zh-CN/attu_search.md

@@ -0,0 +1,47 @@
+---
+id: attu_search.md
+related_key: attu
+summary: Attu, an intuitive GUI for Milvus service management.
+---
+
+# 使用 Attu 进行向量搜索
+
+这个篇文章将描述 Attu 如何进行向量搜索。
+
+## 向量相似度搜索
+
+在常规向量相似度搜索的基础上，还可以进行时间旅行搜索的混合搜索。
+
+### 加载 collection 到缓存中
+
+Milvus 内的所有 CRUD 操作都在内存中执行。在进行向量相似性搜索之前，需要将 collection 加载到内存中。查看[加载 collection](attu_collection.md#Load-a-collection) 了解更多。
+
+![Search Data](../assets/insight_search1.png)
+
+### 设置搜索参数
+
+1. 在 **Choose collection and field** 区域的下拉列表框中选择要搜索的 collection 和向量字段。
+2. 在 **Enter vector value** 字段中，输入一个向量，其维度与要搜索的目标向量具有相同的所选字段维度。
+3. 在 **Set search parameters** 区域，指定索引的特定参数和其他与搜索相关的参数。
+
+![Search Data](../assets/insight_search2.png)
+
+### 带有高级过滤器的混合搜索（可选）
+
+单击 **Advanced Filter**，将出现 **Advanced Filter** 对话框。您可以使用 **AND** 或 **OR** 运算符将多个条件组合成一个复合条件。过滤器表达式会随着条件的任何更改而自动更新。查看 [boolean expression rule](boolean.md) 了解更多。
+
+![Search Data](../assets/insight_search3.png)
+
+### 时间旅行搜索（可选）
+
+Milvus 为所有数据插入和删除操作维护一个时间线。它允许用户在向量搜索中，指定时间戳以检索指定时间点的数据视图。
+
+1. 单击 **Time Travel**，然后在出现的对话框中选择一个时间点。
+
+![Search Data](../assets/insight_search4.png)
+
+2. 在 **TopK** 下拉列表框中，指定要返回的搜索结果数。
+3. 单击 **Search** 根据搜索条件检索最相似的向量。
+
+![Search Data](../assets/insight_search5.png)
+![Search Data](../assets/insight_search6.png)

doc/zh-CN/attu_system.md

@@ -0,0 +1,38 @@
+---
+id: attu_system.md
+related_key: attu
+summary: Attu, an intuitive GUI for Milvus service management.
+---
+
+# 使用 Attu 监控系统状态
+
+此话题讨论如何使用 Attu 监控 Milvus 系统。
+
+## 系统视图
+
+点击左侧导航面板上的**系统视图**图标进入系统视图页面。
+
+![系统视图](../assets/insight_system1.png)
+
+系统视图仪表盘由以下卡片组成：
+
+- **磁盘**： 展示存储空间占用情况。
+- **内存**： 展示内存占用情况。
+- **Qps**： 展示最近 10 次 QPS。每个节点代表一个时间节点。鼠标悬浮上节点可以查看时间节点上 QPS 详情。
+- **延迟**: 展示最近 10 次延迟。每个节点代表一个时间节点。鼠标悬浮上节点可以查看时间节点上延迟详情。
+
+![系统视图](../assets/insight_system2.png)
+![系统视图](../assets/insight_system3.png)
+
+- **拓扑图**： 展示当前运行 Milvus 实例系统结构。点击 Milvus 节点或者 Coordinator 节点后，右侧信息卡片会展示选中节点相关信息。
+- **信息**： 展示选中节点的硬件，系统，配置信息。
+
+## 节点列表视图
+
+所有被同一父层 Coordinator 节点下管理的子节点都将出现在列表中。子节点可以通过 **CPU 核心数**、**CPU 内核使用**、**磁盘使用**及**内存使用**衡量指标排序。
+
+![节点列表视图](../assets/insight_system4.png)
+
+列表右侧是一张迷你版的拓扑图展示算中节点和父层 Coordinator 节点关系。迷你拓扑图下方是相关信息卡片。
+
+点击向下箭头收起节点列表视图。

doc/zh-cn.md

@@ -0,0 +1,24 @@
+# Attu 相关文档
+
+### 安装
+
+- [使用 Docker 安装](./zh-CN/attu_install-docker.md)
+- [使用 Helm 安装](./zh-CN/attu_install-helm.md)
+
+### 管理 Milvus 数据
+
+- [管理 Collection](./zh-CN/attu_collection.md)
+- [管理 Data](./zh-CN/attu_data.md)
+- [管理 Partition](./zh-CN/attu_partition.md)
+- [管理 Index](./zh-CN/attu_index.md)
+
+### 特点
+
+- [结构化匹配](./zh-CN/attu_query.md)
+- [Attu Overview 页面](./zh-CN/attu_overview.md)
+- [向量检索](./zh-CN/attu_search.md)
+- [系统状态](./zh-CN/attu_system.md)
+
+### 问答
+
+- [问答](./zh-CN/attu_faq.md)