DISTRIBUTED COMPUTING

Many successful organisms and biological systems function in a distributed way. Individual ants or bees are not very intelligent. Still, as a colony, they build massive structures and achieve amazing things. The success of such colonies lies in the fact that a large number of organisms try out various options, and the most successful option is copied by the rest of the colony.

Computers can help us -as individuals or as a society- to become even more intelligent, by trying out many alternative scenarios for us in simulations. This way, we can learn much more quickly, than if we were to actually try out every alternative ourselves.

Networks

At the moment, much of the computing power that is available in datacenters and computers around the world is not being used efficiently. By combining computers in large (distributed) networks, this unused computing power can be used to increase the intelligence of the collective brain of our society.

This approach also leads to systems that are more energy efficient. If we avoid running computer systems and databases in vain, we can accomplish much more while using the same amount of energy.

This can be achieved by dynamically distributing computations across a network. Centralized systems have the disadvantage of being very fragile because they often have a (or several) single point of failure. A network of computers is both robust and energy efficient.

Challenges:

> How to build a robust, scalable network

> How do we use the full (computing, storing and communicating) potential of our computer systems?

> How do we deal with network instability?

Keywords:

> energy efficiency

> cloud computing

> networks

> scalability

> robustness