I have a replication setup 1 master and 2 slaves. version of percona is 5.6 and size of database is 550+ GB.
There is some size differences between master and slave1. some tables have small size of .ibd files than master and some tables are partitioned (700+ partition).
Slave was setup using physical files (not using mysqldump).RAM = 500GB
The following details is given of one of big table and it is very important table of our database :-

On master :-

count(*) = 816746410
Total size of table = 261.326690673828 GB
data_length of table = 128.425338745117 GB
index_length of table = 132.901351928711 GB
Rows = 770959074
Avg_row_length = 178

Data_free (Data fragmentation) = 5315231744 (approx 5GB)

on slave :-

count(*) = 816754414
Total size of table = 134.960708618164 GB
data_length of table = 66.542572021484 GB
index_length of table = 68.418136596680 GB
Rows = 390371601
Avg_row_length = 183

Data_free (Data fragmentation) = 5350883328 (approx 5GB)

please provide the solution and how can i check data consistency OR there is no matter on size because count is same. I have knowledge pt-table-checksum but i don’t get any opportunity to use that.

In order to check data consistency through pt-table-checksum you can refer to my post