Just found this very good article about optimizing servers.

The default value of “Reserved block count” takes 5% of usable disk. On a large fs like 813G, it represents about 40G.

These blocks are reserved to the super user to recover from situations where user processes fill up filesystems.
It is absolutely safe to reduce this space to one hundred or so MB.

• check disk space of our filesystem:

$ df -h /dev/sda4
/dev/sda4             813G  418G  354G  55% /home

Before the tuning, we have 354G free.

• check the current number of reserved blocks:

$ tune2fs -l /dev/sda4
...
Reserved block count:     10816865
...

Change this number to 20000.
The blocksize is 4096, 20000 blocks represent about 80MB.

$ tune2fs -r20000 /dev/sda4
...
Reserved block count:     20000
...

• check disk space of our filesystem:

$ df -h /dev/sda4
/dev/sda4             813G  418G  395G  52% /home

We now have a gain of 40GB of free space!

While reading articles about optimization, I read about the accf_http module.

The man page of the module is here, where you can read:

The utility of accf_http is such that a server will not have to context switch several times before performing the initial parsing of the request.

To load this module, use the following command:

# kldload accf_http

To load it at boot time, add the following line in /boot/loader.conf:

accf_http_load="YES"

To check if the module is loaded, use the command kldstat:

# kldstat
Id Refs Address    Size     Name
 1    4 0xc0400000 906518   kernel
 2    1 0xc0d07000 6a32c    acpi.ko
 3    1 0xc5e65000 2000     accf_http.ko