My ideas have been QNAP'd

It has been some considerable time since my last post. Whilst my fingers may not have been typing as much, this does not mean that productive work has ceased. Far from it. I have in recent months encountered a number of options and ideas that will change the scope of my work with the Raspberry Pi.

Within my own home environment, I have been seeking to utilise the RasPi as a solution for a number of different needs; home automation, media centre, games console, personal cloud server, web server, print server, etc. Up front the RasPi has three core strengths from my humble opinion; ease of use, low-cost and energy efficiency at 36W. However, it has some drawbacks as well.

With respect to it's use as a media server, the software solution that I had heard a lot about and had been wanting to use (because it was the one that would work with our main TV) was Plex. However, Plex is not available on the RasPi. With any server that is storing large amounts of data, that is constantly running, having some kind of redundancy in the storage. Let's face it, the time taken to rip several hundred DVD's and CDs to a media server, and the bandwidth required to download others is not insignificant, and if you've done it once, you really don't want to go through it again. Unfortunately, with only USB connections for mass storage options, the RasPi is limited in this aspect, and the more hard drives that you have plug in through a powered hub, or with their own power supply, the less green the RasPi becomes as an option.

Over the last month I have had good fortune to spend some time with fellow geeks who have sorted their home media centres and are using Plex. I've been lucky to receive demonstrations of their systems, and more importantly one of them revealed that he is running his system from a QNAP server. I asked about power consumption and he told me that since replacing an old PC with the QNAP server, their power bills had decreased about $100 (AUD) a year. Given that the QNAP servers have multiple drive bays, mounting multiple hard drives and configuring a RAID array will provide redundancy. They have multiple Gigabit Ethernet ports, half-a dozen USB ports and other features too. What's more, they only draw 40W when in operation (less for sleep mode) and their processors are far more powerful than that of the RasPi.

Awesome!

So, does this mean the end of the Rastaberrian Project? Hell, no! It just simply means that I can refine the scope of my ideas of what to do with the RasPi, and remove the concept of using it as a media, print or personal cloud server. My current thoughts are to place any such server in DMZ, so that it can be accessible externally as well. In considering the RasPi for home automation, my current thoughts are that this should be secured within the home green zone, and if external access is required - facilitated by web services via a secure web page hosted from the DMZ. Use of the RasPi as a games console, or to make an ordinary TV into a Smart TV, I think still holds merit.

Current Theorems - Am I on Track?

Perhaps my previous calculations of power consumption are too simplistic.

In researching green options, I discovered on some forums the use of a 5 port USB powered hub to deliver power to the RPi, and its peripherals, with the RPi delivering communications and data back to the hub.  I had been wondering about the concept of using one hub to power multiple RPis - would it be more efficient and greener?

Both the RPi and the hub run from their own transformers.  The RPi transformer has an input of 240V at 50-60Hz, drawing 150mA, and delivers a DC output of 1A 5VA.  To my limited knowledge of the calculations involved, this means that the input power is: 240 x 0.15 = 36 Watts, and the output is: 5 x 1 = 5 Watts.  This would imply that 31 Watts is lost to the resistance and impedance of the unit, and generates heat in the transformer.

When I look at the input and output of the USB hub's transformer, the input is: 240V at 50-60Hz, drawing 1A, therefore 240 Watts.  The output is 5V 1.5 A, therefore 7.5 Watts.

This would imply that whilst the powered hub supplied 50% more power, it draws 660% more power from the mains, and hence would not be a greener more efficient supply.

However, I know that my knowledge in this area is presently limited - am I using the right calculations and making the right deductions? Should I be calculating and comparing the power consumption of the transformers from their input or output values? Are these constant loads, or the maximum loads of each unit?

Anyone able to help?