Author: Brian Lee Page: 2

On the back of the Qube 2 you'll notice the LCD screen that I mentioned earlier and the input arrows.  You'll also see two 10/100 Ethernet ports.  The primary port is for your WAN connection and the secondary port is for your LAN connection.  This is because you can use the Qube 2 as a NAT or router, plus it has a built in firewall.  Of course you'll have to add a hub or switch to the secondary Ethernet port in order to add more than one computer to the network.  You can also see the PCI slot where you can install a PCI modem or ISDN modem.  Remember that any hardware you install has to be compatible with Linux.  There is also a COM port so you can attach an external analog modem.

Since the Qube is based upon Red Hat Linux, the server is very stable.  You should rarely have to reboot this machine unless something really bad happens.  This stability is great in a business environment where the Qube could be a server to dozens if not hundreds of people.

The newer Qube 3 has either an AMD K6-2 300 MHz or 450 MHz CPU which is fast enough to handle almost anything you can throw at it.  My older Qube 2 has a 250 MHz MIP CPU which is slow as dirt in comparison to the Qube 3.  However, even with this rather slow CPU, it runs most operations very decently and can easily saturate my DSL connection with web pages or emails.  The Qube 2 can handle thousands of emails and millions of web requests a day.  However with the MIP CPU, if your website is made up of primarily dynamic material, i.e.: Perl or PHP, you'll see your performance really drop as your user load goes up.  If your website is primarily static html pages, then this CPU won't even break a sweat under heavy load.  The new Qube 3 is much better at handling dynamic web content.

Qube 2 specs:

• MIPS 250 MHz CPU
• Supports up to 256 MB of 3.3V RAM (expensive and hard to find)
• Supports up to a 30 GB hard drive
• Based on RedHat 5.2 and the 2.0 Linux kernel
• 2 x 100BaseTX network ports
• 1 x PCI slot 
• 1 x Serial port 

You can manage the Qube by using the graphical user interface or you can manage it in terminal mode by telnetting into the server.  Since telnet isn't very secure, many Qube administrators have disabled telnet and installed SSH instead.  Once you open a terminal session on the server, you can do what you like, as long as you know something about Linux.  However, since the Qube is designed for small businesses where you may not have a Linux expert around, most people probably will stick to the graphical interface.  Again, if you know enough to mess around in terminal mode, you probably would build your own Linux box and not mess around with the Qube.

Currently, I have the Qube 2 running as a mail server on my network.  My webserver is still a Windows machine, so you can see that I'm barely taking advantage of the capabilities of the Qube.

You may be thinking that the Qube servers are pretty dumb because a guy who knows Linux could get the same functionality by building their own box.  I don't think this is necessarily true.  First of all, you would need to buy all the hardware necessary for the server.  After installing and configuring all the required components and purchasing some 3rd party software, you could potentially exceed the cost of a Qube.  Also could you build a box that looked as cool as a Qube?  I don't think so!  More important than the cost of the Qube is whether or not you have the time or expertise to build a server as functional and easy to use as the Qube.  If you do, I suggest you go start your own company building those servers because you'll make a ton of money.

Back to Page 1

Forward to Page 3