14.3 SQL Database Support

Database Planning Tips and Caveats

Filr supports the following databases:

  • MySQL 5.0.96 and later

  • Microsoft SQL 2008 R2, 2012, and 2014

If you have a MySQL or MS SQL installation, Micro Focus recommends that you leverage that resource and take advantage of your existing infrastructure.

In an enterprise environment, it is important to ensure that the Filr appliances’ shared storage (the /vashare mount point) is protected by traditional clustering to provide for high availability.

The MySQL or MS SQL server and database that Filr appliances access must also be protected for high availability.

Database Sizing

The amounts shown in “Worksheet 23 - SQL Database” are explained in Table 14-3. As before, you can adjust the numbers shown with solid or small-dotted borders. Numbers in cells with black dashes are calculated based on information you entered previously and you should not modify them directly.

Table 14-3 SQL Server Sizing Guidelines

Item

Recommendations and Guidelines

Planning Worksheet Example

  • / (MySQL appliance)

  • 20 GB fixed

 

  • Database

    /vastorage on MySQL appliance

Use the following guidelines when planning for the SQL database:

  • Allocate 40 GB as a starting amount.

  • Allocate 30 KB per content-searchable file.

  • Allocate 20 KB per name-search-only file.

  • Allocate 20 KB per user.

The worksheet calculates this as follows in both the SQL Database and MySQL Appliance sections:

  1. It includes 40 GB as a base disk amount.

  2. It multiples the total number of content-searchable files by 30 KB (.000030 GB).

  3. It multiples the total number of name-search-only files by 20 KB (.000020 GB).

  4. It multiples the total number of users by 20 KB (.000020 GB).

  • RAM (MySQL appliance)

Use the following guidelines when planning allocating RAM for the MySQL appliance.

  • 8% of /vastorage

  • 60 to 75% of RAM to innodb_buffer_pool.

The worksheet calculates this as follows in the MySQL Appliance section:

  1. It multiples the /vastorage size by .08 to get 8%.

  2. Next, it multiplies the RAM calculation from the first step by .7 to get 70% for the innodb_buffer_pool recommendation.

  • CPUs (MySQL appliance)

  • 4