# Dumpling Architecture and Characteristics

__*Dumpling architecture*__
__*Dumpling Features*__

+ Architecturally, dumpling is a client-side tool that connects to a running TiDB database for data export.
+ This course assumes that there is a TiDB instance at 127.0.0. 1:4000, and that TiDB instance has a root user with no password.
+ The dumpling tool has the following characteristics
  + It supports exporting a variety of data formats, including SQL/CSV.
  + The new table-filter is supported, making it easier to filter data.
  + Export to Amazon S3 Cloud Disk is supported.
  + Optimize for TiDB
-------------------------------------------------------------
It supports exporting a variety of data formats, including SQL/CSV.

The new table-filter is supported, making it easier to filter data.

Export to Amazon S3 Cloud Disk is supported.

More optimizations for TiDB:
    Support configuring TiDB single SQL memory limit.
    Automatic adjustment of TiDB GC time is supported for TiDB v4.0.0 and later.
    Optimized the concurrent export performance of data in a single table using TiDB's hidden column *tidb*rowID.
    For TiDB, you can set the value of tidb_snapshot to specify the point in time to back up the data, thereby ensuring the consistency of the backup, rather than using FLUSH TABLES WITH READ LOCK to ensure the consistency of the backup.