Boffins copy bees to boost web servers

The swarm intelligence of well-organised honey bees could be used to improve the efficiency of web servers, according to new research.

A communications system inspired by bees and developed at the Georgia Institute of Technology helps servers normally devoted to one task to move between tasks as needed.

This load balancing reduces the risk that a website could be overwhelmed with requests and lock out potential users and customers.

The newly developed 'honeybee' method typically improves service by between four and 25 per cent compared to traditional server banks in tests based on real internet traffic.

Craig Tovey, a professor in the H. Milton Stewart School of Industrial and Systems Engineering at Georgia Tech, realised that bees and servers had strikingly similar barriers to efficiency.

"I studied bees for years, waiting for the right application," he said. " When you work with biomimetics you have to look for a close analogy between two systems [and] never a superficial one. And this definitely fit the bill."

Bees tackle their resource allocation problems (i.e. a limited number of bees and unpredictable demands on time and desired location) with a system driven by 'dances'.

The scout bees leave the hive in search of nectar. Once they find a promising spot, they return to the hive 'dance floor' and perform a dance.

The 'direction' of the dance tells the waiting forager bees which direction to fly, the number of 'waggle turns' conveys the distance to the flower patch and the 'length' of the dance conveys the sweetness of the nectar, according to the scientists.

While dancing may not sound like a model of efficiency, Professor Tovey believes that it is optimal for the unpredictable nectar world the bees inhabit.

The system allows the bees to shift seamlessly from one nectar source to a more promising nectar source based on up-to-the-minute conditions.

Tovey and Sunil Nakrani, a computer science colleague visiting from the University of Oxford, set to work translating the dance-based bee strategy for idle internet servers.

Although optimised for 'normal' conditions, such servers are frequently challenged by spikes in demand. To combat this the researchers developed a virtual 'dance floor' for a network of servers.

When one server receives a user request for a certain website, an internal advertisement is placed on the 'dance floor' to attract any available servers.

The ad's duration depends on the demand on the site and how much revenue its users may generate. The longer an ad remains on the 'dance floor', the more power available servers devote to serving the website requests.