New Self-Scheduling Schemes for Internet-Based Grids of Computers

Self-scheduling algorithms are useful tools for achieving load balance in computational systems, in particular, in computational Grids. In this work, we introduce two families of self-scheduling algorithms. The first is defined by methods looking for an explicit form for the chunks distribution function. The second corresponds to methods focusing in derive an explicit form for the rate of variation of the chunks distribution function. From the first family, we select a Quadratic Self-Scheduling (QSS) algorithm. From the second, a new Exponential Self-Scheduling (ESS) algorithm is introduced. QSS and ESS are optimized for working in an Internet-based Grid of Computers involving resources from Spain and Mexico. Performance tests show that QSS and ESS outperform previous selfscheduling algorithms. QSS is found slightly more efficient than ESS.