On Honey Bees and Dynamic Server Allocation in Internet Hosting Centers

  • Authors:
  • Sunil Nakrani;Craig Tovey

  • Affiliations:
  • Computing Laboratory, University of Oxford;School of Industrial and Systems Engineering, Georgia Institute of Technology

  • Venue:
  • Adaptive Behavior - Animals, Animats, Software Agents, Robots, Adaptive Systems
  • Year:
  • 2004

Quantified Score

Hi-index 0.01

Visualization

Abstract

Internet centers host services for e-banks, e-auctions and other clients. Hosting centers then must allocate servers among clients to maximize revenue. The limited number of servers, costs of reallocating servers, and unpredictability of requests make server allocation optimization difficultBased on the many similarities between server and honey bee colony forager allocation, we pro pose a new decentralized honey bee algorithm which dynamically allocates servers to satisfy request loads. We compare it against an omniscient optimality algorithm, a conventional greedy algorithm, and an algorithm that computes omnisciently the optimal static allocation. We evaluate performance on simulated request streams and commercial trace dataOur algorithm performs better than static or greedy for highly variable request loads, but greedy can outperform it under low variability. Honey bee forager allocation, though suboptimal for static food sources, may possess a counterbalancing responsiveness to food source variability.