Grid computing, in short, it the process of connecting multiple computers through a network to use their resources to all achieve a common goal.
Say you want to process 60 units of work (‘units’ is not a standard measurement, in this context it is simply a metaphor) and your machine could process 10 units of work per minute, it would take your machine 6 minutes to complete the task. With grid computing, the 60 units of work are divided to work-sections, and processed by multiple machines. Let’s say that 6 computers are set up in a grid, all of which are capable of working at 10 units per minute. Each machine is assigned to process 10 units of work from the project, so the same task can be achieved in one minute. So if you double the machines, you half the amount of time. Therefore, 12 machines would take 30 seconds, 24 machines, 15 seconds etc.
In effect, with this method of connecting computers, you aren’t increasing the processing power of a ‘machine’ you are simply spreading the workload over many machines, to reduce the processing time. A large project from Stanford University, called Folding@Home, uses computers and game consoles around the world to analyse information. A user simply runs the software on their machine, is then assigned some data to process, processes it on their machine, and then uploads it back to Folding@Home. The more people who do this, the more information that can be processed in a given amount of time.
In theory, if the software supports it, you could do this with a large variety of applications. Adobe already allow you to do this with software such as Adobe After Effects. The rendering time of video can be greatly reduced by spreading the workload. Ultimately, if internet connections can be fast enough, computers all around the world would be able to process vast amounts of information in a short amount of time. Who knows what discoveries this could lead to?
We’d love to hear your thoughts on grid computing. If you want to start a discussion with the Servers Australia community, join us on facebook.
