Oracle database backup DMP file is too big, what method can be split into multiple DMP when backup?

When creating an export, split the export by time or other conditions.
Then write a shell script to cyclically export:
Example:

vim par.pardirectory=DP_DIRdumpfile=t1.dmptables=t1query=t1:" WHERE CREATED_TS >= TO_DATE('20170501 000000', 'YYYY/MM/DD HH24:MI:SS')AND CREATED_TS < TO_DATE('20170601 000000', 'YYYY/MM/DD HH24:MI:SS')"logfile=expdp_test.logcontent=data_only -- only export dataexpdp username/'password' parfile=par.par

Example 2:

Image capture:
httpshttps://www.xifenfei.com/2021/06/expdp-single-table-parallel.html

Another method is to use parallelism. It is a bit dangerous to use parallelism in the production environment. I have not used it, but it is also an idea for your reference:

Image capture:
https://www.it610.com/article/1290850473259114496.htm
I hope it helps you.If helpful, please click to accept

When creating an export, split the export by time or other conditions.
Then write a shell script to cyclically export:
Example:

vim par.pardirectory=DP_DIRdumpfile=t1.dmptables=t1query=t1:" WHERE CREATED_TS >= TO_DATE('20170501 000000', 'YYYY/MM/DD HH24:MI:SS')AND CREATED_TS < TO_DATE('20170601 000000', 'YYYY/MM/DD HH24:MI:SS')"logfile=expdp_test.logcontent=data_only -- only export dataexpdp username/'password' parfile=par.par

Example 2:

Image capture:
httpshttps://www.xifenfei.com/2021/06/expdp-single-table-parallel.html

Another method is to use parallelism. It is a bit dangerous to use parallelism in the production environment. I have not used it, but it is also an idea for your reference:

Image capture:
https://www.it610.com/article/1290850473259114496.htm
I hope it helps you.If helpful, please click to accept

DUMPFILE can replace variables by specifying %U, and multiple dump files can be created. The replacement variables will be from 01 to 99. For example, exp%U.dmp can be dumped into exp01.dmp, exp02.dmp, exp03.dmp and other file names.

Multiple dump files specified with %U, determined by the parameter PARALLEL.
Serving multiple files in a comma-separated list

https://docs.oracle.com/database/121/SUTIL/GUID-A6300021-419F-4C1D-AFF1-38FE1123326B.htm#SUTIL841
https://blog.csdn.net/linsuhangoracle/article/details/118960073

https://docs.oracle.com/cd/B19306_01/server.102/b14215/dp_overview.htm#sthref44

dumpfile %U parameter description

When exporting expdp, there are several parameters that can be specified:

1, filesize----"specify the size of a single dump file

2, %U substitution variable, can create multiple dump files

3, parallel-----" specifies the parallel process started when exporting.

If you want to export multiple files, you can use the above parameters to achieve it. Please note that generally if there is only one table and multiple parallels are started, the performance improvement is not obvious.

expdp mobile/mobile directory=DATA_PUMP_DIR dumpfile=UserBoss_20160619_25%U.dmp parallel=4 filesize=500M