• Business Technology Consulting Graduate Blog
• Consulting Blog

What Ever Happened to Distributed Computing?

What ever happened to distributed computing? I was having a look through the technology pages of the BBC News website today and came across an old article from just over 5 years ago proclaiming that distributed computing would be the next big thing to transform the Internet…and it would be transforming it very soon! So if 5 years ago it was set to turbo-charge your computer using the reservoirs of power connected to the Internet, why aren’t we seeing the fruits of this revolution now?

Distributed computing is the means of harnessing the power of idle computers connected to a network by distributing the processing between them, allowing multiple machines to accomplish what a single, stand-alone super-computer could achieve for a fraction of the cost. The power of distributed computing can be amplified when you choose to split up and process data across the largest network in the world, the Internet – this is known as grid computing. Classic examples of previous success stories for distributed computing have been SETI@home, where radio-telescope data is analysed to find evidence of intelligent signals from space, and Folding@home, where machines are used to replicate protein folding to find cures for diseases and understand biophysical systems.

So, with all this potential and the power it could have harnessed with some 500 million devices connected to the Internet worldwide in 2003, what ever happened to distributed computing? Well, the truth is that distributed computing is no longer the ‘hot topic’, meaning it doesn’t come up in conversation anymore and people therefore assume it is no longer around, when in reality it has actually been absorbed into a much bigger hot topic; Cloud Computing.

You might be wondering why, before being absorbed into Cloud Computing, the average user didn’t utilise distributed computing… Quite simple really – the average user didn’t need to utilise distributed computing. Moore’s Law dictates that every other year the number of transistors that can be inexpensively placed on a circuit board will double, and with processing power being directly linked to this law, new technology provided us with all the power needed in a single machine.

The advancements in hardware explains why the average user didn’t need distributed computing, so if it is not a prehistoric theory shelved in the archives of the London Science Museum, where is it being used today? The answer is data centres; warehouses packed with technologies that handle and store of massive amounts of data, distributing the processing across hugely powerful servers, and as you might have guessed, the next ‘big thing’ to make use of data centres is Cloud Computing.

With Cloud Computing, the data and corresponding application to utilise that data is stored in a data centre (or a cloud, as it is not tangible to a user of the service), and when the Cloud Service is used to access the data, all the processing is completed in the data centre with the output being subsequently streamed via the internet to the user’s substantially less powerful device. If the processing required is too much for one server in the data warehouse, it can distribute the load to surrounding servers, and voila; you have a fully functioning Cloud Computing service where a network of servers completing distributed processing within a data centre provide a service to a third party device via the internet, effectively outsourcing your in-house distributed computing to a CO2 efficient cloud.

Alternatively, you could look at it from a different angle; if your machine has fallen victim to a virus and joined the ranks in the botnet army would you think then that distributed computing has lost favour in the IT world, or just that it has been exploited by hackers on a global scale?