|
|
| Web Servers |
Our web servers are
custom-built industrial machines designed for a 24/7 web serving environment. All of our
servers are equipped with dual redundant 450-watt power supplies, hot swap Fujitsu drives
and force-filtered cooling systems. Our Network Operations Center (NOC) is equipped
with standby machines, industry-leading Liebert uninterruptible power supplies (UPSs) and
a temperature-controlled environment, all located in a secure, monitored facility. |
|
Force Filtered Cooling
All the drives, motherboards, and power
supplies in our NOC are in cases with a positive pressure filtered-air environment. Two
large fans pull filtered air into the protective case, and the components within are
cooled by fans which circulate this purified air. The constant introduction of clean air
into the case creates a positive pressure environment ensuring that dust and particles
remain out. |
| Hot Swap Fujitsu Drives |
The drives and drive
bays of our servers are all constructed from high-grade aluminum, and rest in a Shock
Mounted Drive Cage, which adds to the durability of the hardware. We back up our servers
nightly. If the industrial grade drive were to fail, all we would have to do is copy the
backup (never more than 24 hours old) onto a standby machine which is already on-line. The
site would be up and running within a matter of minutes. |
| Redundant Hot Swap Power Supplies |
Each server has
dual-redundant hot swap power supplies. If one of these power supplies were to fail, the
server would not go down because each power supply is capable of running the server by
itself, indefinitely. Meanwhile, alarms would alert our technicians to the problem; they
would restore redundancy with one of our spares and then service the failed supply unit.
During all this, sites on our network would never be down. |
| Standby Servers |
We keep spares on-line
of all CPU configurations. If the server were to experience a hardware failure we would
literally turn a key, grab the handle on the drive, pull it out, and insert it into an
identical standby CPU. We would reboot the second machine and the server would be up and
running again in a matter of minute |
|
|