Merge branch 'preview' of github.com:milvus-io/web-content into preview

v2.4.x/Variables.json

@@ -1,6 +1,6 @@
 {
-  "milvus_release_version": "2.4.17",
-  "milvus_release_tag": "2.4.17",
+  "milvus_release_version": "2.5.x",
+  "milvus_release_tag": "2.5.x",
   "milvus_deb_name": "milvus_2.2.0-1_amd64",
   "milvus_rpm_name": "milvus-2.2.0-1.el7.x86_64",
   "milvus_python_sdk_version": "2.4.x",

v2.4.x/format.js

@@ -0,0 +1,107 @@
+const fs = require('fs');
+const {
+  join
+} = require('path');
+
+const formatMenuStructure = list => {
+  const newList = list.map(v => {
+    const {
+      id,
+      title,
+      isMenu = false,
+      outLink = '',
+      order = 0,
+      label1,
+      label2,
+      label3,
+    } = v;
+
+    const parentId = label3 || label2 || label1 || '';
+    const parentIds = [label1, label2, label3].filter(v => !!v);
+    const level = [label1, label2, label3].filter(v => !!v).length + 1;
+
+    const baseInfo = {
+      id: id,
+      label: title,
+    }
+
+    if (isMenu) {
+      baseInfo.isMenu = true;
+    }
+
+    if (outLink) {
+      baseInfo.externalLink = outLink;
+    }
+
+    return {
+      ...baseInfo,
+      parentId,
+      parentIds,
+      level,
+      order,
+      children: [],
+    };
+  });
+
+  newList.sort((x, y) => y.level - x.level);
+
+  const resultList = newList.slice();
+
+  newList.forEach(v => {
+    const {
+      parentId
+    } = v;
+    const parentIndex = resultList.findIndex(v => v.id === parentId);
+    if (parentIndex !== -1) {
+
+      let childInfo = {
+        label: v.label,
+        id: v.id,
+
+        // href: v.href,
+        // parentId: parentId,
+        // parentIds: v.parentIds,
+        // level: v.level,
+        order: v.order,
+
+      }
+
+      if(v.isMenu){
+        childInfo.isMenu = true;
+      }
+
+      if(v.externalLink){
+        childInfo.externalLink = v.externalLink
+      }
+
+      childInfo.children = v.children;
+
+      resultList[parentIndex].children.push(childInfo);
+    }
+  });
+
+  return resultList.filter(v => v.level === 1).map(v => ({
+    label: v.label,
+    id: v.id,
+    isMenu: v.isMenu,
+    externalLink: v.externalLink,
+    // href: v.href,
+    // parentId: parentId,
+    // parentIds: v.parentIds,
+    // level: v.level,
+    order: v.order,
+    children: v.children,
+  }));
+};
+
+const formatMenu = () => {
+
+
+  const menuPath = join(__dirname, 'site/en/menuStructure/en.json')
+
+  const menu = fs.readFileSync(menuPath, 'utf-8');
+  const newMenuStructure = formatMenuStructure(JSON.parse(menu).menuList);
+
+  fs.writeFileSync(menuPath, JSON.stringify(newMenuStructure));
+};
+formatMenu();

v2.4.x/site/en/about/overview.md

@@ -80,7 +80,7 @@ In addition to primitive data types, Milvus supports various advanced data types
 - [Sparse Vectors](sparse_vector.md)
 - [Binary Vectors](index-vector-fields.md)
 - [JSON Support](use-json-fields.md)
-- [Array Support](array_data_type.md)
+- [Array Support](array.md)
 - [Distance Metrics](metric.md)
 
 ### Acceleration

v2.4.x/site/en/about/roadmap.md

@@ -15,35 +15,35 @@ Welcome to the Milvus Roadmap! Join us on our continuous journey to enhance and
     <thead>
         <tr>
             <th>Category</th>
-            <th>Milvus 2.4.0 (Recently Achieved)</th>
-            <th>Milvus 2.5.0 (Upcoming in Mid-CY24)</th>
-            <th>Future Roadmap (Milvus 3.0 Expected within CY24)</th>
+            <th>Milvus 2.5.0 (Achieved in recent releases)</th>
+            <th>Next Release (Middle of CY25)</th>
+            <th>Future Roadmap (Within 1 year)</th>
         </tr>
     </thead>
     <tbody>
         <tr>
-            <td><strong>AI-developer Friendly</strong><br/><i>A developer-friendly technology stack, enhanced with the latest AI innovations</i></td>
-            <td><strong>Multi-Vectors & Hybrid Search</strong><br/><i>Framework for multiplex recall and fusion</i><br/><br/><strong>GPU Index Acceleration</strong><br/><i>Support for higher QPS and faster index creation</i><br/><br/><strong>Model Library in PyMilvus</strong><br/><i>Integrated embedding models for Milvus</i></td>
-            <td><strong>Sparse Vector (GA)</strong><br/><i>Local feature extraction and keyword search</i><br/><br/><strong>Milvus Lite (GA)</strong><br/><i>A lightweight, in-memory version of Milvus</i><br/><br/><strong>Embedding Models Gallery</strong><br/><i>Support for image and multi-modal embeddings and reranker models in model libraries</i></td>
-            <td><strong>Original Data-In and Data-Out</strong><br/><i>Support for Blob data types</i><br/><br/><strong>Data Clustering</strong><br/><i>Data co-locality</i><br/><br/><strong>Scenario-oriented Vector Search</strong><br/><i>e.g. Multi-target search & NN filtering</i><br/><br/><strong>Support Embedding & Reranker Endpoint</strong></td>
+            <td><strong>AI-Driven Unstructured Data Processing</strong><br/><i>Strengthening the ability to process and analyze unstructured data using AI models and advanced technologies.</i></td>
+            <td><strong>Full Text Search</strong><br/><i>Support full text search with Sparse-BM25. The new API accepts text as input and automatically generates sparse vector inside Milvus</i><br/><br/><strong>Sparse Vector(GA)</strong><br/><i>Support efficient storage and indexing method for sparse vector</i><br/></td>
+            <td><strong>Data-In and Data-Out</strong><br/><i>Support major model services to ingest original data</i><br/><br/><strong>Advanced Reranker</strong><br/><i>Support model-based rerankers and user-defined scoring function</i><br/><br/><strong>JSON Enhancement</strong><br/><i>JSON indexing and parsing to accelerate processing</i></td>
+            <td><strong>Original Data-In and Data-Out</strong><br/><i>Support Blob and url reference to process original data</i><br/><br/><strong>Support More Data Types</strong><br/><i>e.g. Datetime, Map, GIS</i><br/><br/><strong>Support Tensors</strong><br/><i>Support list of vectors, typical usage like Colbert, Copali etc.</i></td>
         </tr>
         <tr>
-            <td><strong>Rich Functionality</strong><br/><i>Enhanced retrieval and data management features</i></td>
-            <td><strong>Support for FP16, BF16 Datatypes</strong><br/><i>These ML datatypes can help reduce memory usage</i><br/><br/><strong>Grouping Search</strong><br/><i>Aggregate split embeddings</i><br/><br/><strong>Fuzzy Match and Inverted Index</strong><br/><i>Support for fuzzy matching and inverted indexing for scalar types like varchar and int</i></td>
-            <td><strong>Inverted Index for Array & JSON</strong><br/><i>Indexing for array and partial support JSON</i><br/><br/><strong>Bitset Index</strong><br/><i>Improved execution speed and future data aggregation</i><br/><br/><strong>Truncate Collection</strong><br/><i>Allows data clearance while preserving metadata</i><br/><br/><strong>Support for NULL and Default Values</strong></td>
-            <td><strong>Support for More Datatypes</strong><br/><i>e.g. Datetime, GIS</i><br/><br/><strong>Advanced Text Filtering</strong><br/><i>e.g. Match Phrase</i><br/><br/><strong>Primary Key Deduplication</strong></td>
+            <td><strong>Search Quality & Performance</strong><br/><i>Deliver accurate, relevant, and fast results by optimizing architecture, algorithms and APIs</i></td>
+            <td><strong>Text Match Function</strong><br/><i>Quickly filter keywords/tokens in text/varchar</i><br/><br/><strong>Grouping Search Enhancement</strong><br/><i>Introduce group_size and add group by support in hybrid search</i><br/><br/><strong>Bitmap Index & Inverted Index</strong><br/><i>Accelerate filtering on tags</i></td>
+            <td><strong>Advanced Match</strong><br/><i>e.g. Match Phrase, Fuzzy Match, and more tokenizers</i><br/><br/><strong>Aggregations</strong><br/><i>Scalar field aggregations, e.g. min, max, count, distinct.</i><br/></td>
+            <td><strong>Partial Update</strong><br/><i>Support updates to a specific field's value</i><br/><br/><strong>Sorting Capability</strong><br/><i>Sort by scalar fields during execution</i><br/><br/><strong>Support Data Clustering</strong><br/><i>Data co-locality</i></td>
         </tr>
         <tr>
-            <td><strong>Cost Efficiency & Architecture</strong><br/><i>Advanced systems emphasizing stability, cost efficiency, scalability, and performance</i></td>
-            <td><strong>Support for More Collections/Partitions</strong><br/><i>Handles over 10,000 collections in smaller clusters</i><br/><br/><strong>Mmap Optimization</strong><br/><i>Balances reduced memory consumption with latency</i><br/><br/><strong>Bulk Insert Optimazation</strong><br/><i>Simplifies importing large datasets</i></td>
-            <td><strong>Lazy Load</strong><br/><i>Data is loaded on-demand through read operations</i><br/><br/><strong>Major Compaction</strong><br/><i>Re-distributes data based on configuration to enhance read performance</i><br/><br/><strong>Mmap for Growing Data</strong><br/><i>Mmap files for expanding data segments</i></td>
-            <td><strong>Memory Control</strong><br/><i>Reduces out-of-memory issues and provides global memory management</i><br/><br/><strong>LogNode Introduction</strong><br/><i>Ensures global consistency and addresses the single-point bottleneck in root coordination</i><br/><br/><strong>Storage Format V2</strong><br/><i>Universal format design lays the groundwork for disk-based data access</i></td>
+            <td><strong>Rich Functionality & Management</strong><br/><i>Developer-friendly and robust data management features</i></td>
+            <td><strong>Support CSV files in data import</strong><br/><i>Bulkinsert supports CSV format</i><br/><br/><strong>Support Null and Default Value</strong><br/><i>Null and Default types make importing data from other DBMS easier</i><br/><br/><strong>Milvus WebUI (Beta)</strong><br/><i>Visual management tools for DBAs</i></td>
+            <td><strong>Primary Key Deduplication</strong><br/><i>By using the global pk index</i><br/><br/><strong>Online Schema Change</strong><br/><i>e.g. Add/delete field, modify varchar length</i><br/><br/><strong>Data Versioning & Restore</strong><br/><i>Support data versioning by snapshot</i></td>
+            <td><strong>Rust and C++ SDK</strong><br/><i>Support more clients</i><br/><br/><strong>Support UDF </strong><br/><i>User-defined function</i></td>
         </tr>
         <tr>
-            <td><strong>Enterprise Ready</strong><br/><i>Designed to meet the needs of enterprise production environments</i></td>
-            <td><strong>Milvus CDC</strong><br/><i>Capability for data replication</i><br/><br/><strong>Accesslog Enhancement</strong><br/><i>Detailed recording for audit and tracing</i></td>
-            <td><strong>New Resource Group</strong><br/><i>Enhanced resource management</i><br/><br/><strong>Storage Hook</strong><br/><i>Support for Bring Your Own Key (BYOK) encryption</i></td>
-            <td><strong>Dynamic Replica Number Adjustment</strong><br/><i>Facilitates dynamic changes to the number of replicas</i><br/><br/><strong>Dynamic Schema Modification</strong><br/><i>e.g., Add/delete fields, modify varchar lengths</i><br/><br/><strong>Rust and C# SDKs</strong></td>
+            <td><strong>Cost Efficiency & Achitecture</strong><br/><i>State-of-the-art systems, prioritizing stability, cost-efficiency and scalability </i></td>
+            <td><strong>Load by Field</strong><br/><i>Choose part of collection to load</i><br/><br/><strong>Memory Optimization</strong><br/><i>Reduce OOM and load enhancement</i><br/><br/><strong>Streaming Node (Beta)</strong><br/><i>Provide global consistency and solve performance bottleneck on root coordinator</i><br/><br/><strong>Storage Format V2 (Beta)</strong><br/><i>Universal formats design and foundation for disk-based data access</i><br/><br/><strong>Clustering Compaction</strong><br/><i>Data re-distribution based on configuration to accelerate read performance</i></td>
+            <td><strong>Lazy Load</strong><br/><i>Load can be initiated by the first read operation without explicitly calling load()</i><br/><br/><strong>Tiered Storage</strong><br/><i>Support hot and cold storage for cost optimization</i><br/><br/><strong>Release by Field</strong><br/><i>Release part of collection to reduce memory usage</i><br/><br/><strong>Streaming Node (GA)</strong><br/><i>Process streaming data and simplify the architecture</i></td>
+            <td><strong>Remove dependencies</strong><br/><i>Reduce or eliminate dependencies on external components like pulsar, etcd</i><br/><br/><strong>Merge coord logic into MixCoord</strong><br/><i>Simplify the architecture</i></td>
         </tr>
     </tbody>
 </table>

v2.4.x/site/en/adminGuide/configure-docker.md

@@ -16,10 +16,10 @@ In current release, all parameters take effect only after Milvus restarts.
 
 ## Download a configuration file
 
-[Download](https://raw.githubusercontent.com/milvus-io/milvus/v2.4.17/configs/milvus.yaml) `milvus.yaml` directly or with the following command.
+[Download](https://raw.githubusercontent.com/milvus-io/milvus/v2.5.x/configs/milvus.yaml) `milvus.yaml` directly or with the following command.
 
 ```
-$ wget https://raw.githubusercontent.com/milvus-io/milvus/v2.4.17/configs/milvus.yaml
+$ wget https://raw.githubusercontent.com/milvus-io/milvus/v2.5.x/configs/milvus.yaml
 ```
 
 ## Modify the configuration file
@@ -178,13 +178,13 @@ Sorted by:
 
 ## Download an installation file
 
-Download the installation file for Milvus [standalone](https://github.com/milvus-io/milvus/releases/download/v2.4.17/milvus-standalone-docker-compose.yml), and save it as `docker-compose.yml`.
+Download the installation file for Milvus [standalone](https://github.com/milvus-io/milvus/releases/download/v2.5.x/milvus-standalone-docker-compose.yml), and save it as `docker-compose.yml`.
 
 You can also simply run the following command.
 
 ```
 # For Milvus standalone
-$ wget https://github.com/milvus-io/milvus/releases/download/v2.4.17/milvus-standalone-docker-compose.yml -O docker-compose.yml
+$ wget https://github.com/milvus-io/milvus/releases/download/v2.5.x/milvus-standalone-docker-compose.yml -O docker-compose.yml
 ```
 
 ## Modify the installation file

v2.4.x/site/en/adminGuide/operator.md

@@ -12,7 +12,7 @@ Milvus cluster depends on components including object storage, etcd, and Pulsar.
 
 This topic assumes that you have deployed Milvus Operator.
 
-<div class="alert note">See <a href="https://milvus.io/docs/v2.4.17/install_cluster-milvusoperator.md">Deploy Milvus Operator</a> for more information. </div>
+<div class="alert note">See <a href="https://milvus.io/docs/v2.5.x/install_cluster-milvusoperator.md">Deploy Milvus Operator</a> for more information. </div>
 
 You need to specify a configuration file for using Milvus Operator to start a Milvus cluster.
 

v2.4.x/site/en/adminGuide/resource_group.md

@@ -56,7 +56,7 @@ All code samples on this page are in PyMilvus 2.4.9. Upgrade your PyMilvus insta
 
     To create a resource group, run the following after you connect to a Milvus instance. The following snippet assumes that `default` is the alias of your Milvus connection.
 
-    ```Python
+    ```python
     import pymilvus
 
     # A resource group name should be a string of 1 to 255 characters, starting with a letter or an underscore (_) and containing only numbers, letters, and underscores (_).
@@ -80,7 +80,7 @@ All code samples on this page are in PyMilvus 2.4.9. Upgrade your PyMilvus insta
 
     To view the list of resource groups in a Milvus instance, do as follows:
 
-    ```Python
+    ```python
     rgs = utility.list_resource_groups(using='default')
     print(f"Resource group list: {rgs}")
 
@@ -91,7 +91,7 @@ All code samples on this page are in PyMilvus 2.4.9. Upgrade your PyMilvus insta
 
     You can have Milvus describe a resource group in concern as follows:
 
-    ```Python
+    ```python
     info = utility.describe_resource_group(name, using="default")
     print(f"Resource group description: {info}")
 
@@ -110,7 +110,7 @@ All code samples on this page are in PyMilvus 2.4.9. Upgrade your PyMilvus insta
     Assuming there are currently 1 QueryNodes in the **__default_resource_group** of the cluster, and we want to transfer one node into created **rg**.
     `update_resource_groups` ensures atomicity for multiple configuration changes, so no intermediate states will be visible to Milvus. 
 
-    ```Python
+    ```python
     source = '__default_resource_group'
     target = 'rg'
     expected_num_nodes_in_default = 0
@@ -138,7 +138,7 @@ All code samples on this page are in PyMilvus 2.4.9. Upgrade your PyMilvus insta
 
     Once there are query nodes in a resource group, you can load collections to this resource group. The following snippet assumes that a collection named `demo` already exists.
 
-    ```Python
+    ```python
     from pymilvus import Collection
 
     collection = Collection('demo')
@@ -154,7 +154,7 @@ All code samples on this page are in PyMilvus 2.4.9. Upgrade your PyMilvus insta
 
     Also, you can just load a partition into a resource group and have its replicas distributed among several resource groups. The following assumes that a collection named `Books` already exists and it has a partition named `Novels`.
 
-    ```Python
+    ```python
     collection = Collection("Books")
 
     # Use the load method of a collection to load one of its partition
@@ -173,7 +173,7 @@ All code samples on this page are in PyMilvus 2.4.9. Upgrade your PyMilvus insta
 
     Milvus uses [replicas](replica.md) to achieve load-balancing among [segments](glossary.md#Segment) distributed across several query nodes. You can move certain replicas of a collection from one resource group to another as follows:
 
-    ```Python
+    ```python
     source = '__default_resource_group'
     target = 'rg'
     collection_name = 'c'
@@ -192,7 +192,7 @@ All code samples on this page are in PyMilvus 2.4.9. Upgrade your PyMilvus insta
 
     You can drop a resource group that hold no query node (`limits.node_num = 0`) at any time. In this guide, resource group `rg` now has one query node. You need to change the configuration `limits.node_num` of resource group into zero first.
 
-    ```Python
+    ```python
     try:
         utility.update_resource_groups({
             "rg": utility.ResourceGroupConfig(
@@ -218,7 +218,7 @@ Here is a good practice for managing QueryNodes in a cloud environment:
     Additionally, if we strictly enforce the constraint `sum(.requests.nodeNum) <= queryNodeNum`, we can precisely control the assignment of QueryNodes in the cluster. Let's assume there is currently only one QueryNode in the cluster and initialize the cluster.
     Here is an example setup:
 
-    ```Python
+    ```python
     from pymilvus import utility
     from pymilvus.client.types import ResourceGroupConfig
 
@@ -261,15 +261,15 @@ Here is a good practice for managing QueryNodes in a cloud environment:
 2. Cluster scale out
 
     Assuming we have the following scaling function:
-    ```Python
+    ```python
 
     def scale_to(node_num: int):
         # scale the querynode number in Milvus into node_num.
         pass
     ```
 
     We can use the API to scale a specific resource group to a designated number of QueryNodes without affecting any other resource groups.
-    ```Python
+    ```python
     # scale rg1 into 3 nodes, rg2 into 1 nodes
     utility.update_resource_groups({
         "rg1": ResourceGroupConfig(
@@ -293,7 +293,7 @@ Here is a good practice for managing QueryNodes in a cloud environment:
 
     Similarly, we can establish scaling-in rules that prioritize selecting QueryNodes from **__pending_nodes** resource group. This information can be obtained through the `describe_resource_group` API. Achieving the goal of scaling-in specified resource group.
 
-    ```Python
+    ```python
     # scale rg1 from 3 nodes into 2 nodes
     utility.update_resource_groups({
         "rg1": ResourceGroupConfig(