A couple of weeks ago I put the following into my signature:
Browse the internet and help fight cancer, HIV/AIDs and water pollution by donating your CPU's excess processing time to basic scientific research.
It's a little non-specific as to what I'm talking about, so I thought I write up an explanation.
Grid computing is a form of computation. Unlike conventional computing that uses only one machine; grid computing uses multiple resources from many locations in tandem to reach a solution to a single problem. That can be achieved a number of ways, but the most common way to do it is for a central hub to take a large problem and break it apart into very tiny pieces. The hub will then farm out the pieces to other computers, and after each piece is resolved separately, stitch back together the results.
A lot of problems can be resolved this way. The most famous project using this method is the SETI@home program. Several telescopes in this program are continually scanning the skies for incoming electromagnetic waves. The researchers then break up all of this data and farm it out to thousands of participants whose computers scan the data for abnormal patterns during times when the computer's CPU is idle. For example, the short span of idle time between when the users opens up an article, takes the time to read it, and then continues browsing. The goal is to find patterns that may indicate extraterrestrial life.
Likewise my link is another favorite grid computing program of mine that is run by IBM. In it, IBM works with various research groups to tackle large computational problems that are scientific in nature. To give an example, several of the longer running projects -- like FightHIV@home, Help Stop TB and Outsmart Ebola -- have mapped the HIV virus or Ebola and are computing the interactions between the virus and tens of millions of compounds. The goal is to find a compound that interacts or "binds" to the virus, which then can be used by scientists to design more effective drugs that will interact better with the virus.
I find this type of technology fascinating. While on one hand we can build giant computers that cost millions of dollars, we've found a way to harness the otherwise-wasted idle time of small computers scattered across the world in a fashion that can compete with the big league computers.
Browse the internet and help fight cancer, HIV/AIDs and water pollution by donating your CPU's excess processing time to basic scientific research.
It's a little non-specific as to what I'm talking about, so I thought I write up an explanation.
Grid computing is a form of computation. Unlike conventional computing that uses only one machine; grid computing uses multiple resources from many locations in tandem to reach a solution to a single problem. That can be achieved a number of ways, but the most common way to do it is for a central hub to take a large problem and break it apart into very tiny pieces. The hub will then farm out the pieces to other computers, and after each piece is resolved separately, stitch back together the results.
A lot of problems can be resolved this way. The most famous project using this method is the SETI@home program. Several telescopes in this program are continually scanning the skies for incoming electromagnetic waves. The researchers then break up all of this data and farm it out to thousands of participants whose computers scan the data for abnormal patterns during times when the computer's CPU is idle. For example, the short span of idle time between when the users opens up an article, takes the time to read it, and then continues browsing. The goal is to find patterns that may indicate extraterrestrial life.
Likewise my link is another favorite grid computing program of mine that is run by IBM. In it, IBM works with various research groups to tackle large computational problems that are scientific in nature. To give an example, several of the longer running projects -- like FightHIV@home, Help Stop TB and Outsmart Ebola -- have mapped the HIV virus or Ebola and are computing the interactions between the virus and tens of millions of compounds. The goal is to find a compound that interacts or "binds" to the virus, which then can be used by scientists to design more effective drugs that will interact better with the virus.
I find this type of technology fascinating. While on one hand we can build giant computers that cost millions of dollars, we've found a way to harness the otherwise-wasted idle time of small computers scattered across the world in a fashion that can compete with the big league computers.