Difference between consumer and server components

[adamwri]

I might be in the market to pick up a new server shortly so I've been looking into components and checking out some costs. I'm a little put off by the idea of spending a small fortune on components which, in comparison to their consumer models, don't appear to be any better on the surface.

What's the obvious differences between components and their intended purpose? Is there really that much difference that would be noticed? Seems crazy to fork out £2,000 on a machine which could be had for half that!

 

[Rilot]

The main difference is reliability. If you look at a Proliant DL380p Gen8 for example you have redundant PSUs, redundant PCI-E cages, lots of memory channels, ECC memory, Xeon E5 processors with lots of memory channels, hot-plug drives in two separate domains, etc etc.
But the big thing you're paying for is supportability. Many ISVs will only support their applications on server hardware.

 

[joeyjojo]

I'm a little put off by the idea of spending a small fortune on components which, in comparison to their consumer models, don't appear to be any better on the surface.

If this is the case then you don't need server components. Simple really.

 

[Judgeneo]

As mentioned, support is the big thing your a paying for.

If you don't really want that, then there isn't much stopping you from buying consumer grade equipment.

 

[Quartz]

Another thing to consider is the space requirements. Space is expensive. Servers are often supplied ready to be fitted into a rack, thereby consuming a minimum of space.