Creating incremental syncs（创建增量同步）

Although it is possible to derive APPEND-only datasets in a pipeline from Data Connection syncs that are configured as SNAPSHOT transactions, most of the benefits of incremental pipelines come from applying incremental end-to-end. This means that data syncs into Foundry should consist of APPEND transactions that only bring new data into the system. An added benefit of configuring incremental syncs is that they minimize load on the source system and can reduce data storage requirements.

Most datasets synced from source systems consist of files synced from a file system, or extracts from a database or data warehouse configured using a JDBC source type. The following guides walk you through how to configure incremental syncs for these source types:

• Optimize file-based append syncs
• Incremental JDBC syncs

---

中文翻译

创建增量同步

虽然可以从配置为SNAPSHOT（快照）事务的Data Connection同步中推导出仅包含APPEND（追加）操作的流水线数据集，但增量流水线的大部分优势来自于端到端的增量应用。这意味着同步到Foundry的数据应由仅将新数据引入系统的APPEND（追加）事务组成。配置增量同步的另一个好处是，它们能最小化对源系统的负载，并可能降低数据存储需求。

从源系统同步的大多数数据集包括从文件系统同步的文件，或使用JDBC源类型配置的数据库或数据仓库的提取数据。以下指南将引导您如何为这些源类型配置增量同步：

• 优化基于文件的追加同步
• 增量JDBC同步