The cluster-concepts document #436
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| An IoTDB cluster consists of three types of nodes (processes): **ConfigNode** (the main node), **DataNode**, and **AINode**, as detailed below: | ||
| - **ConfigNode:** ConfigNodes store cluster configurations, database metadata, the routing information of time series' schema and data. They also monitor cluster nodes and conduct load balancing. All ConfigNodes maintain full mutual backups, as shown in the figure with ConfigNode-1, ConfigNode-2, and ConfigNode-3. ConfigNodes do not directly handle client read or write requests. Instead, they guide the distribution of time series' schema and data within the cluster using a series of [load balancing algorithms](https://iotdb.apache.org/UserGuide/latest/Technical-Insider/Cluster-data-partitioning.html). | ||
| - **DataNode:** DataNodes are responsible for reading and writing time series' schema and data. Each DataNode can accept client read and write requests and provide corresponding services, as illustrated with DataNode-1, DataNode-2, and DataNode-3 in the above figure. When a DataNode receives client requests, it can process them directly or forward them if it has the relevant routing information cached locally. Otherwise, it queries the ConfigNode for routing details and caches the information to improve the efficiency of subsequent requests. | ||
| - **AINode:** AINodes interact with ConfigNodes and DataNodes to extend IoTDB's capabilities for machine learning analysis on time series data. They support registering pre-trained machine learning models from external sources and performing time series analysis tasks using simple SQL statements on specified data. This process integrates model creation, management, and inference within the database engine. Currently, the system provides built-in algorithms or custom models for common time series analysis scenarios, such as forecasting and anomaly detection. |
Contributor
There was a problem hiding this comment.
machine learning analysis-> data intelligence analysis
custom models -> self-training models
| IoTDB 集群包括三种节点（进程），**ConfigNode**（管理节点），**DataNode**（数据节点）和 **AINode**（分析节点），如下所示： | ||
| - **ConfigNode**：存储集群的配置信息、数据库的元数据、时间序列元数据和数据的路由信息，监控集群节点并实施负载均衡，所有 ConfigNode 之间互为全量备份，如上图中的 ConfigNode-1，ConfigNode-2 和 ConfigNode-3 所示。ConfigNode 不直接接收客户端读写请求，它会通过一系列[负载均衡算法](https://iotdb.apache.org/zh/UserGuide/latest/Technical-Insider/Cluster-data-partitioning.html)对集群中元数据和数据的分布提供指导。 | ||
| - **DataNode**：负责时间序列元数据和数据的读写，每个 DataNode 都能接收客户端读写请求并提供相应服务，如上图中的 DataNode-1，DataNode-2 和 DataNode-3 所示。接收客户端读写请求时，若 DataNode 缓存有对应的路由信息，它能直接在本地执行或是转发这些请求；否则它会向 ConfigNode 询问并缓存路由信息，以加速后续请求的服务效率。 | ||
| - **AINode**：负责与 ConfigNode 和 DataNode 交互来扩展 IoTDB 集群对时间序列进行机器学习分析的能力，支持从外部引入已有机器学习模型进行注册，并使用注册的模型在指定时序数据上通过简单 SQL 语句完成时序分析任务的过程，将模型的创建、管理及推理融合在数据库引擎中。目前已提供常见时序分析场景（例如预测与异常检测）的机器学习算法或自研模型。 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
As the title describes.