How to maintain the oldest JIRA database?

[Soft Admin]

Dear supporter: 
 For the hard drive space issue,what is your best recommendation to maintain the oldest JIRA database. 
I mean if I run JIRA in 5 years,and just want to keep the closer 1 year data. However ,the other 4 years data will be archive and compress or delete.
My enviroment is
Database info:
==================================================================
| version | 5.3.11-MariaDB-log |
| version_comment | Source distribution |
| version_compile_machine | x86_64 exit
| storage_engine | InnoDB
| table_type | InnoDB 
===================================================================
My Server & system info
8core cpu; 16G RAM; 300G Hard drive 
CentOS release 6.4 
===================================================================
My JIRA version is v6.3.15#6346-sha1:dbc023d

Best regards,

ERNIE HONG

104GROUP

[Nic Brough -Adaptavist-]

Set up a second instance of JIRA, export the main JIRA data to XML and import the projects you want to "archive" in the new instance.  In the second JIRA, have a read-only permission scheme.  Once the data is archived in the second JIRA, delete the projects from the first one.

Do not try to do this at a database level - you need a complete understanding of the structure of the data, and it is far more simple to rely on the export/import stuff which has that built in.