Biology inspired computing
I read an article in Communications of the ACM on Biology inspired networking(it’s premium content online 😦 -> http://cacm.acm.org/magazines/2011/6/108641-biology-inspired-networking/fulltext).
Biology inspired computing is a branch that tries to understand the problems in biological systems, how they’re solved and how this could be applied to the field of computer science. It’s one among the many inter-disciplinary fields that have developed in recent times.
Biology inspired computing is a very young field that is still to mature. One of the key issues of this field is bridging the collaborators i.e. computer scientists and biologists since both adopt a different approach to their work and finding problems that are of interest to both biologists and computer scientists alike.
A recent research(biology related networking) involves application of neurological development of a fruit fly to distributed peer to peer networks. The project, headed by Ziv Bar Joseph of Carnegie Mellon University(CMU), aims at creating a new networking algorithm that will replace the existing algorithm for managing distributed networks.
In a fruit fly’s nervous system, the cells are unaware of the connection between themselves and thus allocates some cells as leaders so that all cells can be connected-very similar to the concept of peer to peer networks but more robust and unlike anything that exists in today’s distributed networks.
The fruit fly cells send out chemical signals once elected as the Sensory Organ Precusor(SOP) that prevents the neighbouring cells from becoming the SOP. This continues till all cells are either SOP’s or neigbours of SOPs. However, unlike in a distributed network system where a node is elected as the leader as a function of number of connected but as a function of time.
This however is much slower than the existing algorithm but much more efficient. It’s suitable for application in sensor networks where speed is not so important as efficiency since the entire operation of sensor nodes depends on this network’s existence.
Other possible areas where biology could inspire solutions include fault tolerance systems in which the node that is bound to crash/fail could be backed up more often-methods to determine could perhaps be used to make the network robust and highly available.
This was an eye opening article-I knew computational biology existed which involved anaylysis of existing data but wasn’t very interested in it. But, I never thought that other fields could contribute to advancement of computer science! I’m defintely excited about the relevance of inter-disciplinary learning that advances both fields and thos related to it too! Gotta find out more about Biology inspired computation.