As the economy continues to recover, many companies are seeking ways to streamline efficiency without spending a fortune. One area of efficiency that has seen a significant boost recently is low cost computing.
Currently, the UK-based Raspberry PI Foundation is getting ready to release what is being hailed as the $35 computer. Unlike many past attempts at sub-$100 PCs, this PC can run Ubuntu, stream 1080i video and provide connectivity video output to PC monitors and TVs of all types. The Raspberry PI seems like it might be like a super-low cost thin client, when in reality it's a fully functional standalone PC.
Now, as exciting as it is to have access to computers that range from $25 for 128MB to $35 for the full 256 MB unit, there are some concerns that need to be addressed. And one of these concerns is whether you’ll be able to readily buy one of these computers.
In order for super-cheap computing to take off, it would help to have more than one group selling a reliable product. Thankfully, this appears to be on the horizon, even if it's not always at a $35 price point.
For example, the CuBox may cost more at $135, but it offers the end user more out of the box. First, you can pre-order yours now instead of waiting for the Raspberry PI to offer pre-ordering themselves. Secondly, you can get more RAM for your unit with a rock solid 1GB vs the 128-256MB seen with the Raspberry PI.
The more expensive CuBox will be the more likely candidate for developers looking to get their hands on a low cost PC than the Raspberry PI. Why is this? Because the Raspberry PI is (mostly) going to be offered to non-profit organizations while the CuBox will be available on a first come first served basis.
The big question that remains in my mind: who will end up being the buyer for these low cost PCs? The CuBox is marketed as a development platform, while the Raspberry PI is designed to meet the computing needs of schools and other non-profits. In either case, I don't expect to see low cost computers making their way into big box stores anytime soon.
Access to distributed computing
The concept behind distributed computing isn't anything new. Giving away CPU cycles has been a big thing among computer enthusiasts for years with programs such as SETI@home, in addition to other distributed computing concepts.
But it's really not worthwhile for anyone to participate in if it means running a full-sized PC 24/7. Plus, it also means that you won’t be able to run a distributed computing network in-house since it would require a ton of additional resources.
This means that if you have a few rooms full of these sub-100 dollar PCs running distributed computing software with a common goal in mind, private companies would able to create their own virtual supercomputers on the cheap.
Even better, there's no need for soliciting assistance from the outside world such as we've seen from other distributed computing projects. So any large distributed computing projects that a company would rather keep in-house would definitely be workable.
How much computer do we need?
Over the years, I've seen many people fall victim to the “newer is better” PC marketing nonsense put forth by many PC manufacturers. And while it's important for some intensive tasks to have access to the best resources possible, not every computer out there needs 4 Gbs of RAM with the latest processor.
Depending on the duties assigned to the workstation in question, it's entirely possible that a lower powered, low cost PC might be a great match for most common computing needs.